# MiniMax M3

MiniMax M3 הוא מודל מבית MiniMax עם משקולות פתוחות, שהושק בתאריך 2026-06-01. חלון ההקשר הוא 1,000,000 טוקנים.

- מפתח: MiniMax
- Id: minimaxai/minimax-m3
- לפי דיווחים: 2026-03-17T15:12:20.000Z
- הושק: 2026-06-01T00:00:00.000Z
- זיהוי: 2026-09-17T00:52:24.374Z
  - [קישור לראיה](https://models.dev/#minimax-m3)
- סטטוס: הושק
- גרסת המשקולות: 2026-06-12
  - [קישור לראיה](https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3/paths-info/d95e30cb912fe3021bb9c2ba66a8ce60b6c937a9)
- קנוני: https://alexgetman.com/he/models/minimax/minimax-m3-677/

## שאלות

### מהו חלון ההקשר של MiniMax M3?

חלון ההקשר הוא 1,000,000 טוקנים.

### האם המשקולות של MiniMax M3 זמינות לציבור?

כן. המשקולות פורסמו על ידי MiniMax.

### מתי הושק MiniMax M3?

בתאריך 2026-06-01.

## ציר זמן

- 2026-11-05T17:47:34.513Z · [GitHub · MiniMax-AI/minimax-code · code](https://github.com/MiniMax-AI/minimax-code/blob/3c21163c85a9684030cb147bfef48c6284fa3c46/packages/agent-modules/context-manager/src/settings.ts)
> afety margin.
>  */
> export function computeCompactionTriggerAt(input: {
>   modelId?: string;
>   contextWindow: number;
>   perTurnMaxTokens: number;
>   reserveTokens: number;
>   safetyMarginTokens: number;
> }): number {
>   if (
>     input.modelId === 'MiniMax-M3' &&
>     (input.contextWindow === 512_000 || input.contextWindow === 1_000_000)
>   ) {
>     return Math.floor(input.contextWindow * 0.9);
>   }
>   const perTurnMaxTokens =
>     Number.isFinite(input.perTurnMaxTokens) && input.perTurnMaxTokens > 0
>       ? input.perTurnMaxTokens
>       : 0;
>   const reserve = Math.max(input.reserveTokens, perTurnMaxTokens + input.safetyMarginTokens);
>   return Math.max(1, input.contextWindow - reserve);
> }
> 
האזכור הזה בקוד אינו מאשר מזהה מודל שנמצא בשימוש.
- 2026-11-15T15:09:30.906Z · [GitHub search · MiniMax · pulls · ikawrakow/ik_llama.cpp · @mb8565](https://github.com/ikawrakow/ik_llama.cpp/pull/2885)
> SWIGLU_OAI / SITU when it would fall back MiniMax-M3 GGUFs abort while the graph is built, with default flags, when the dense or shared-expert up/gate
- 2026-11-15T15:09:30.906Z · [GitHub search · MiniMax · pulls · im-dongseon/wikihub · @im-dongseon](https://github.com/im-dongseon/wikihub/pull/224)
> | 구현 리뷰 | `ollama-cloud/kimi-k2.7-code`, `ollama-cloud/minimax-m3` (`--model` 직접, subagent 프롬프트 없음) |
- 2026-11-15T15:09:30.906Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @waqahmed-amd-fi](https://github.com/sgl-project/sglang/pull/39994)
> that fails `_has_dense_kv_rows()` is made contiguous first; the MiniMax-M3 and Unlimited-OCR v2 index-K sources are already dense. The two paths differ only on slots a
- 2026-11-15T15:09:56.939Z · [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/54a45d9f4d853314b61407387674f484b03730c6)
> Model: minimax/MiniMax-M3
- 2026-11-15T15:09:56.939Z · [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/6e8ec57325d12347e7644f60d3196974db97a6e8)
> Model: minimax/MiniMax-M3
- 2026-11-15T15:09:56.939Z · [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/7f8bd83f8f4363133c7d274782359c49a805aa21)
> Model: minimax/MiniMax-M3
- 2026-11-15T15:09:56.939Z · [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/d6eb81e4d4c2ea0f15ef32b08480b92df24c3795)
> Model: minimax/MiniMax-M3
- 2026-11-15T15:09:56.939Z · [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/ec9335a877acfe5e8bf1a7aeef0384a9d2f7037f)
> Model: minimax/MiniMax-M3
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · AtomFlow-AI/MoleCode · @octo-patch](https://github.com/AtomFlow-AI/MoleCode/pull/2)
> LLMClient with global and China endpoint presets and MiniMax-M3 / MiniMax-M2.7 model support.
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · Cmochance/codex-app-transfer · @octo-patch](https://github.com/Cmochance/codex-app-transfer/pull/614)
> Anthropic-compatible MiniMax presets now offer Flash first and MiniMax-M3, with Flash's documented 1,000,000-token context and 524,288-token output limit. The provider form filters obsolete MiniMax text
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · anchapin/fluxion · @anchapin](https://github.com/anchapin/fluxion/pull/4723)
> Worker: MiniMax-M3 (flagship), 60/60 steps, 23,999 tokens out. Brief was Goldfish-reviewed before dispatch (9 findings, all applied) and narrowed per failover reflection from the prior
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · anchapin/planar-nexus · @anchapin](https://github.com/anchapin/planar-nexus/pull/2884)
> Worker: minimax/MiniMax-M3 (flagship), 60/60 steps, 3,395,187 tokens. Supervisor completed verification, tmp-profiling cleanup, lint fix, commit, rebase, and PR.
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · anymorph-ai/Claudable · @octo-patch](https://github.com/anymorph-ai/Claudable/pull/86)
> has no MiniMax provider. Add the provider with MiniMax-M3.1-Flash as the default, MiniMax-M3 as the other ordinary choice, and Anthropic-compatible Claude Agent SDK execution.
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · b1tninja/continuous · @b1tninja](https://github.com/b1tninja/continuous/pull/513)
> type=T<|sep|>...`, encoding_k3.py), not kimi-tokens. MiniMax M3: its own `minimax-m3` row (namespaced `]<]minimax[>[<invoke name=...>`). Both `textDialect: none` + "not yet recoverable"; the dialect enum is untouched
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @Kangyan-Zhou](https://github.com/sgl-project/sglang/pull/49392)
> to engine-side rendering. This adds them, following the MiniMax-M3 renderer.
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @AndreasKaratzas](https://github.com/vllm-project/vllm/pull/63486)
> tests move to `tests/models/deepseek_v41/amd_specialized_decode/`, next to Kimi-K3's and MiniMax-M3's
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @esmeetu](https://github.com/vllm-project/vllm/pull/66203)
> the specialized decode requires DSpark. Kimi-K3/K3.1 on NVIDIA, MiniMax-M3, and DeepSeek-V4-Pro/V4.1 on AMD (`--monokernel`) run without speculative decoding, so they belong in a standalone feature
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @ZhengGong-amd](https://github.com/vllm-project/vllm/pull/66695)
> Kimi-K3, Kimi-Linear-48B and DeepSeek-R1 reach it; DeepSeek-V4/V4.1, GLM-5.3/-Flash, MiniMax-M3, Qwen3.x, gpt-oss, gemma-4 and Llama do not. With Kimi-K3's `--max-num-batched-tokens 8192` recipe a context chunk reaches
- 2026-11-15T15:25:28.610Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @ZhengGong-amd](https://github.com/vllm-project/vllm/pull/66706)
> EMULATION]`, `VLLM_ROCM_USE_AITER` off by default): the Qwen3.5-397B-A17B, Qwen3.8-2.4T-A95B, MiniMax-M3 and GLM-5.3 Quark MX-FP4 checkpoints run `OCP_MXQuantizationEmulationTritonExperts` on MI355X unless AITER is on. `QuarkW4A8MXFp4FP8._apply_emulation` and the
- 2026-11-15T15:27:30.324Z · [GitHub search · MiniMax · commits · Mutinho/agents-dashboard · @unknown](https://github.com/Mutinho/agents-dashboard/commit/82553191868d8e82a5a8382e490c5325829d6053)
> - dev.md model=openrouter/minimax/minimax-m3; opencode.jsonc + seed.ts coherentes
- 2026-11-15T15:27:30.324Z · [GitHub search · MiniMax · commits · angelstreet/mycodebox · @unknown](https://github.com/angelstreet/mycodebox/commit/850e1c87a2339923ebfca0aae97153ace7602d2f)
> models cache, MiniMax-M3), a prompt box (attach, paste or drop images/.md) and preselects every choice of the last create.
- 2026-11-15T15:27:30.324Z · [GitHub search · MiniMax · commits · inference-sim/blis-substrate · @sriumcp](https://github.com/inference-sim/blis-substrate/commit/0ae394c26d3059536dd56a6fa18c2fed919a55a0)
> (minimax_m3/nvidia/model.py:529-538), and every MiniMax-M3 header stores it F32 [128, 6144] on the
- 2026-11-15T15:27:30.324Z · [GitHub search · MiniMax · commits · inference-sim/blis-substrate · @sriumcp](https://github.com/inference-sim/blis-substrate/commit/83a95d7a483ccd1b7c07548df484653f545dd44b)
> eagle3, dflash and dspark names alone; known-bad, a MiniMax-M3 config without num_mtp_modules is refused, and a model whose config declares no MTP layers has
- 2026-11-15T15:27:30.324Z · [GitHub search · MiniMax · commits · use-agent-os/agent-os · @andreapn](https://github.com/use-agent-os/agent-os/commit/4c9ba83f609465703b30f3547ac526663ae88bb8)
> whose target is no longer a shipped default (minimax-m3,
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · DennisLoska/skills · @DennisLoska](https://github.com/DennisLoska/skills/pull/13)
> expensive OpenCode Go Pro models (`deepseek-v4-pro`, `kimi-k3.1-pro`, `mimo-v2.6-pro`, `minimax-m3`, `qwen3.8-max`).
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · Inferact/llm-multimodal · @liangjuf](https://github.com/Inferact/llm-multimodal/pull/4)
> MiniMax-M3/M3.1 image and video sizing follows the M3.1 sizing spec — the **aligned long-side resize** (`compute_aligned_size` of the public M3.1 processor) — not the Qwen-style
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @pranav-nvidia](https://github.com/NVIDIA/TensorRT-LLM/pull/20848)
> - Ran the MiniMax-M3 MSA unit tests (`tests/unittest/_torch/attention/sparse/msa/`, `tests/unittest/_torch/models/test_minimax_m3.py`, `tests/unittest/disaggregated/test_minimax_m3_kv_transfer.py`) on SM120 and GB300 on the merged head.
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @yijingl-nvidia](https://github.com/NVIDIA/TensorRT-LLM/pull/21869)
> MiniMax-M3 VL read the image grid back to the host in each vision layer (about 33 syncs per image). It now reads it once per
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @mecworks](https://github.com/NousResearch/hermes-agent/pull/152431)
> hand-added. On top of that, `_MINIMAX_MODELS` ended at `MiniMax-M3`.
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · Pb314314/vllm · @Pb314314](https://github.com/Pb314314/vllm/pull/21)
> them for DFlash and for the Qwen3, Gemma4, MiniMax-M3 and Muse Spark DSpark drafts, but not for the DeepSeek-V4.1 DSpark draft, which keeps all of
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · ROCm/ATOM · @rbrugaro-amd](https://github.com/ROCm/ATOM/pull/2582)
> `DSparkMarkovHead.sample_block` in `dspark_draft.py`, Kimi-K3's loop; V4, V4.1 and MiniMax-M3 keep their own, as their confidence head reads each position's W1[x], which the sampled op does
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @AndreasKaratzas](https://github.com/ROCm/aiter/pull/7053)
> M 6-8 and gate_up at M 33-128, and MiniMax-M3 qkv at TP4/TP8 for most M from 6-9 to 128. In gsm8k serving of Qwen3-8B at
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · Scoheart/arca · @Scoheart](https://github.com/Scoheart/arca/pull/294)
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · a-mansari/tarik · @a-mansari](https://github.com/a-mansari/tarik/pull/209)
> 1. No per-LLM price: only MiniMax-M3 is priced, so extras show tokens, not cost.
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · ashhart/TensorFold · @ashhart](https://github.com/ashhart/TensorFold/pull/1749)
> every registered CUDA launch's image, and #1388 registered MiniMax-M3 launches against `minimax_m3` and `minimax_m3_attn`, which the test's lookup in `zig/tests/cuda/contract_limits.zig` did not know.
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · batipang/CC8 · @batipang](https://github.com/batipang/CC8/pull/493)
> Java 格式、mypy 27 个根模块通过；Android release 编译及 lint 通过，使用原签名。固定访客文字 MiniMax-M3 API 实调成功，没有发送主人历史或调用设备资源。
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · can1357/oh-my-pi · @mikucat0309](https://github.com/can1357/oh-my-pi/pull/18101)
> | `opencode-go/minimax-m3` (Anthropic Messages) | `/inference/go/anthropic/v1/messages` | `ENDPOINT_OK` |
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · dzianisv/hermes-agent · @dzianisv](https://github.com/dzianisv/hermes-agent/pull/40)
> Similar models: `minimax/minimax-m3.1-flash`, `minimax/minimax-m3`, `minimax/minimax-m2.7`
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · jsavoy93/trading-bot · @jsavoy93](https://github.com/jsavoy93/trading-bot/pull/141)
> - the dispatch examples' model changed from MiniMax-M3 to gpt-5.6-sol;
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · leehack/llamadart · @leehack](https://github.com/leehack/llamadart/pull/1910)
> - **Affected-family real-model/artifact evidence:** MiniMax M3 weights unavailable; primary upstream emissions from the llama.cpp `cd959bae9` probe on the vendored template plus durable fixtures
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · nexu-io/open-design · @orionp13](https://github.com/nexu-io/open-design/pull/8196)
> …\|bedrock` | `200` — live catalogue (kimi-k3, glm-5.3, minimax-m3, deepseek-v4-flash, …) |
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · smg-project/smg · @hello-alexmcc](https://github.com/smg-project/smg/pull/5243)
> | MiniMax-M3: preprocess stage median, jpeg-1080p | 6.54 ms (6.51-6.82) | 4.50 ms (-31 %) | `examples/mm_baseline`, release, parent and head built with one cargo
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @zixi-qi](https://github.com/vllm-project/vllm/pull/65217)
> LiLiCorr and Ripple DFlash2 use its list; Gemma4, MiniMax-M3 and Muse Spark DSpark have their own. Other draft models keep every weight.
- 2026-11-15T15:42:52.059Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @AndreasKaratzas](https://github.com/vllm-project/vllm/pull/66370)
> at capture, and the addresses do not change. MiniMax-M3 AMD binds capacity views and checks staleness by `data_ptr`. The DSv4/DSv4.1 and Muse Spark AMD mono
- 2026-11-15T15:44:54.280Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/859c8e8d58f06953e59453bcb53e0b74cae8f697)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · Eco-Sphere/kv-cache-calculator · @Gongdayao](https://github.com/Eco-Sphere/kv-cache-calculator/pull/14)
> a hard cap for every model (previously only minimax-m3 dropped it), include the indexer row in the per-device breakdown, and strip "sequences" from it
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · EvolvingLMMs-Lab/lmms-eval · @Luodian](https://github.com/EvolvingLMMs-Lab/lmms-eval/pull/1628)
> --model_args model=minimax-m3,base_url=http://127.0.0.1:8710/v1,api_key=EMPTY,pass_video_url=True,num_concurrent=4,video_target_frames=32 \
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · MandalorianBatman/austere-axi · @MandalorianBatman](https://github.com/MandalorianBatman/austere-axi/pull/10)
> MINIMAX_API_KEY=… npm run bench # models: minimax-m3 (API) + claude-sonnet (claude CLI login)
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @WeiHaocheng](https://github.com/NVIDIA/TensorRT-LLM/pull/21712)
> - `test_minimax_m3_requires_explicit_opt_in` → `test_minimax_m3_is_claimed_in_every_serving_role`: MiniMax-M3 is now claimed under the default mode too.
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · QuantumNous/new-api · @anhaohighness](https://github.com/QuantumNous/new-api/pull/8281)
> - Applicable types: relay；`POST /v1/chat/completions`，OpenAI 兼容渠道，原报告模型 `MiniMax-M3`，OpenAI → OpenAI、流式、非透传。本次使用本地合成响应；frontend / deployment 不适用，usage 仅核对返回统计。
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · ROCm/ATOM · @XiaobingSuper](https://github.com/ROCm/ATOM/pull/3052)
> main is rejected as predating #5809. On gfx1250, MiniMax-M3 has no gluon fallback and fails model startup.
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @lirui927](https://github.com/ROCm/aiter/pull/7549)
> - in a MiniMax-M3 TP4 server trace the `direct_copy float` after the dense varlen FMHA is gone; it was about 3 us per call plus
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · TencentQQGYLab/AppAgent · @octo-patch](https://github.com/TencentQQGYLab/AppAgent/pull/127)
> AppAgent needs a MiniMax provider that defaults to MiniMax-M3.1-Flash and offers MiniMax-M3 while preserving explicit saved model selections and screenshot inputs.
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · alibaba/open-code-review · @octo-patch](https://github.com/alibaba/open-code-review/pull/1784)
> currently offer M3 and legacy M2 variants. Suggest `MiniMax-M3.1-Flash` first and `MiniMax-M3` second for both native regions, while keeping saved models and explicit legacy/custom IDs
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · enkryptai/helm-charts · @vvemulapalli11](https://github.com/enkryptai/helm-charts/pull/174)
> LiteLLM then drops `tools`, `tool_choice` and `response_format` from MiniMax-M3 requests. gemma's request parameters do not change. Enkrypt confirmed that on-prem red team will route to
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · gudlyf/blynky · @gudlyf](https://github.com/gudlyf/blynky/pull/32)
> The default `OLLAMA_MODEL` was `minimax-m3:cloud`. Ollama's docs say `:cloud` models run on Ollama's servers, so an install that left `OLLAMA_MODEL` at its default sent every
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · jundot/omlx · @jundot](https://github.com/jundot/omlx/pull/5083)
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · leehack/llamadart · @leehack](https://github.com/leehack/llamadart/pull/1894)
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/24)
> > Model `MiniMax-M3` on `router.ai.developerz.ai` via `native`, funded by `account_key`
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @hnyls2002](https://github.com/sgl-project/sglang/pull/48776)
> dict reached only the templates that read it (MiniMax-M3). Now the request normalizer also reads it for every model:
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · vllm-project/semantic-router · @octo-patch](https://github.com/vllm-project/semantic-router/pull/5108)
> Reason: Add MiniMax-M3.1-Flash as the first direct MiniMax text suggestion, followed by MiniMax-M3, while preserving explicit legacy/custom model IDs and existing non-MiniMax bindings.
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · vllm-project/vllm-ascend · @yingyingzizi](https://github.com/vllm-project/vllm-ascend/pull/18428)
> [Refactor][Model] Use CANN GBSA for MiniMax-M3 sparse attention
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · vllm-project/vllm-ascend · @ella1107](https://github.com/vllm-project/vllm-ascend/pull/21460)
> DeepSeek V3/R1, Qwen3 series, GLM-5 series, Gemma4, Kimi-K2/K3, MiniMax-M3, and others, ensuring the localized documentation is up-to-date with the latest supported models.
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · vllm-project/vllm-gaudi · @pawel-olejniczak](https://github.com/vllm-project/vllm-gaudi/pull/1965)
> to import, and so does model-registry inspection of MiniMax-M3. Past the import, the model fails to build on `vision_config` and `rope_theta`.
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @vorapolsiloai](https://github.com/vllm-project/vllm/pull/65311)
> idle. This adds an opt-in sequence-parallel path for MiniMax-M3's vision encoder for single-image batches: the image's patch rows are split across the TP ranks (`VLLM_MINIMAX_M3_VIT_SP`,
- 2026-11-15T15:58:07.252Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @tomasruizt](https://github.com/vllm-project/vllm/pull/66521)
> Gemma 4, Muse Spark). With top-p only (GLM-5.3, MiniMax-M3, Mistral Large 4, Nemotron 3.5), real verify rows are TBD faster on B300. Models without top-k/top-p
- 2026-11-15T15:59:52.324Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/d8247ccdde55069887af433da8b6d0caa2f32989)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T15:59:52.324Z · [GitHub search · MiniMax · commits · vllm-project/vllm-ascend · @FORFuture37](https://github.com/vllm-project/vllm-ascend/commit/514a647be60f6044bb542f7be2529619559b1a15)
> - Served model `minimax-m3` from `/mnt/share/weight/MiniMax-M3-MXFP8`.
- 2026-11-15T15:59:52.324Z · [GitHub search · MiniMax · commits · vllm-project/vllm-ascend · @dianWang325](https://github.com/vllm-project/vllm-ascend/commit/8767c6c401e60e1f9c8dc00b8821955192ce5737)
> **MiniMax-M3 affected-model follow-up (A3):** W8A8, TP8/PP2/DP1 with EP,
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · LMCache/LMCache · @rebel-jinhwan](https://github.com/LMCache/LMCache/pull/6027)
> the drafter's KV, and a stage holding only MiniMax-M3's dense layers has no sparse-attention indexer cache.
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · Plocr/dsh-commandcode · @Mou7s](https://github.com/Plocr/dsh-commandcode/pull/1)
> exist. `deepseek/deepseek-v4-pro` is really `off/high/max`, `zai-org/GLM-5.2` is `high/max`, `MiniMaxAI/MiniMax-M3` is `low/medium/high`, and models such as `moonshotai/Kimi-K2.7-Code` have no effort control at all.
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/35)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/36)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/37)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/40)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · dimagi/open-chat-studio · @claude[bot]](https://github.com/dimagi/open-chat-studio/pull/4989)
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · flytiger-eco/sglang-for-sail · @plusbang](https://github.com/flytiger-eco/sglang-for-sail/pull/256)
> - b235b60d1d feat(minimax-m3): support PPU CUDA SAIL MSA attention via plugin hooks
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · lamtuanvu/pacer · @lamtuanvu](https://github.com/lamtuanvu/pacer/pull/833)
> "Model offered" for a dead default ("…; try opencode-go/minimax-m3") and skips the test task.
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/25)
> > Model `MiniMax-M3` on `router.ai.developerz.ai` via `native`, funded by `account_key`
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · sergiodelgadotech/kotlin-saas-template · @renovate[bot]](https://github.com/sergiodelgadotech/kotlin-saas-template/pull/181)
> integrations with one-click Google and Microsoft OAuth login, **MiniMax-M3 & MiniMax-M3.1-Flash** multimodal LLM serving with DSpark speculative decoding and 8-GPU / multi-node execution, **Qwen3.8-27B-FP8** single-A100
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @michaelzhang-ai](https://github.com/sgl-project/sglang/pull/46944)
> graphs, the aiter FP8 GEMM backend, and the minimax-m3 parsers. GSM8K runs through sgl-eval on the full split (1319 questions, 8192 max tokens, temperature 1.0,
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · vercel-labs/fx · @fazxes](https://github.com/vercel-labs/fx/pull/1658)
- 2026-11-15T16:13:16.406Z · [GitHub search · MiniMax · pulls · xorbitsai/inference · @David-Wu1119](https://github.com/xorbitsai/inference/pull/6051)
> fix(llm): parse MiniMax-M3 tool calls in the namespaced format
- 2026-11-15T16:16:10.409Z · [GitHub search · MiniMax · commits · IHUI-INF-AI/IHUI-AI · @unknown](https://github.com/IHUI-INF-AI/IHUI-AI/commit/f2fa36983ffc765c285553755f814b73bd9ddc37)
> 实测 hy4-preview-x / minimax-m3 / hunyuan-2.0-thinking 三个模型,上游回
- 2026-11-15T16:16:10.409Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/6e6cc9a0ac0f774cdd83713031afc1c1d444c688)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T16:16:10.409Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/a172d71984aaafda89af8184b1191975c9352b5c)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T16:16:10.409Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/e71210da62fdf0044b0b5ad16dd7fb703c26ca3e)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T16:16:10.409Z · [GitHub search · MiniMax · commits · evorule/evo-agent · @EvoRuleLab](https://github.com/evorule/evo-agent/commit/9f31f3cd5fd1348283c61efcae54f1f02ed2465b)
> definition's model (the toml default, MiniMax-M3) - the selector
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · AMD-AGI/Primus · @lhzhang333](https://github.com/AMD-AGI/Primus/pull/1418)
> with Primus-Turbo `dev/mi455x_v4` 126a485b (AMD-AGI/Primus-Turbo#632's head). The DeepSeek-V3, MiniMax-M3 and Llama-3.1-8B examples default to it and no longer need `TURBO_DIR`; READMEs updated. GPT-OSS stays on
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · FelipeGazapina/pinned · @FelipeGazapina](https://github.com/FelipeGazapina/pinned/pull/108)
> with the costly words first. `opus-fast` is high; `minimax-m3` is mid (it read as `mini`). Test: `delegate::tests::a_costly_word_in_a_model_id_outranks_a_cheap_one`.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · Kotlin/kotlin-jupyter-libraries · @renovate[bot]](https://github.com/Kotlin/kotlin-jupyter-libraries/pull/774)
> integrations with one-click Google and Microsoft OAuth login, **MiniMax-M3 & MiniMax-M3.1-Flash** multimodal LLM serving with DSpark speculative decoding and 8-GPU / multi-node execution, **Qwen3.8-27B-FP8** single-A100
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @peihu-nv](https://github.com/NVIDIA/TensorRT-LLM/pull/20629)
> This PR speeds up the MiniMax-M3 NVFP4 context (prefill) worker on Rubin (sm_107). Changes are on by default and engage on sm_107 only; each change
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · OpenOSINT/OpenOSINT · @octo-patch](https://github.com/OpenOSINT/OpenOSINT/pull/40)
> MiniMax configuration. This adds global/CN MiniMax selection with MiniMax-M3.1-Flash as the default and MiniMax-M3 as the other recommended text model, while preserving explicit custom and
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · ai-dynamo/dynamo · @furionw](https://github.com/ai-dynamo/dynamo/pull/17027)
> Known limitation: legacy MiniMax-M3 mixed reasoning/answer deltas still lack ordering metadata; the existing review thread remains open.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · anchapin/fluxion · @anchapin](https://github.com/anchapin/fluxion/pull/4651)
> Worker: minimax/MiniMax-M3 (flagship), 60/60 steps, 4,288,895 tokens.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · arcships/zdr-monitor · @cunninghamcard-bit](https://github.com/arcships/zdr-monitor/pull/1166)
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · awesome-dsh-plugin/awesome-dsh-plugin · @Wahwayhale](https://github.com/awesome-dsh-plugin/awesome-dsh-plugin/pull/12938)
> GLM-5.3-Flash / GLM-5.3 / DeepSeek-V4.1-Flash / DeepSeek-V4-Pro / MiniMax-M3.1-Flash / MiniMax-M3 / Kimi-K3.1-Pro / Kimi-K3), each carrying its credit-cost multiplier.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/ai-army · @developerz-ai[bot]](https://github.com/developerz-ai/ai-army/pull/100)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/ai-gaming-agent-mcp · @developerz-ai[bot]](https://github.com/developerz-ai/ai-gaming-agent-mcp/pull/27)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/ai-gaming-agent-mcp · @developerz-ai[bot]](https://github.com/developerz-ai/ai-gaming-agent-mcp/pull/29)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/claude-task-master · @developerz-ai[bot]](https://github.com/developerz-ai/claude-task-master/pull/199)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/claude-task-master · @developerz-ai[bot]](https://github.com/developerz-ai/claude-task-master/pull/201)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/db-mcp-gateway · @developerz-ai[bot]](https://github.com/developerz-ai/db-mcp-gateway/pull/277)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · developerz-ai/open-mmorpg · @developerz-ai[bot]](https://github.com/developerz-ai/open-mmorpg/pull/115)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · diegoami/malpaco · @diegoami](https://github.com/diegoami/malpaco/pull/474)
> links, and the fullscreen Help close-control fix. Owner-launched MiniMax-M3 thinking milestone review returned AGREE with no findings. The owner explicitly authorized publication for web and
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · gi-dellav/zerostack · @octo-patch](https://github.com/gi-dellav/zerostack/pull/320)
> - Default all four direct regional recipes to MiniMax-M3.1-Flash, retain MiniMax-M3 as the alternative, and document current prices and protocol-specific thinking/effort settings.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · hao-aaron/vllm · @hao-aaron](https://github.com/hao-aaron/vllm/pull/15)
> `DeepseekV32ForCausalLM`) onto `AutoWeightsLoader`, following the pattern of vllm-project/vllm#65904 (MiniMax-M3).
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3239)
> DFlash, DFlash2 (GQA and MLA), K3 DSpark and MiniMax-M3 DSpark draft models have identical module trees, parameter and buffer names, shapes and dtypes, attention sliding
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3241)
> (9), MI355 (5) and GB200 (7); over a MiniMax-M3-only change it still keeps only the MiniMax-M3 eval.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · onyx-dot-app/onyx · @octo-patch](https://github.com/onyx-dot-app/onyx/pull/12918)
> - Add MiniMax-M3 and MiniMax-M2.7 model metadata,
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · sidalit/freetoken-snap · @renovate[bot]](https://github.com/sidalit/freetoken-snap/pull/1)
> - **`qwen3vl` / `qwen3vlmoe`, `muse-glimmer` and `minimax-m3` GGUF models with a fitting `*mmproj*.gguf` in their folder now build the vision tower.** `--text-model-only` keeps them text-only
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @Pb314314](https://github.com/vllm-project/vllm/pull/65972)
> have experts of 25-66M parameters (DeepSeek-V3/V4, Kimi-K2/K3, GLM-5, MiniMax-M3, Qwen3.8-2.4T), i.e. the GLM / M3 side.
- 2026-11-15T16:29:02.877Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @hao-aaron](https://github.com/vllm-project/vllm/pull/66640)
> `DeepseekV32ForCausalLM`) onto `AutoWeightsLoader`, following the pattern of #65904 (MiniMax-M3).
- 2026-11-15T16:31:21.562Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/1100b65a33c347e1f5b5c805f3859a510949eb5f)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T16:31:21.562Z · [GitHub search · MiniMax · commits · gildrb/eukhe · @gildrb](https://github.com/gildrb/eukhe/commit/d845ed27d0d3760982fac83de2845023336801ff)
> - MiniMax-M3 on minimax and minimax-cn answers in at most 128000 tokens:
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · AMD-AGI/Primus-Turbo · @botaohu001](https://github.com/AMD-AGI/Primus-Turbo/pull/627)
> | MiniMax-M3 avg_m 512 default path | 4c74631 = dev/mi455x_v4 = untuned arm = this PR with a tuning argument | 18/18 |
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · AtomFlow-AI/MoleCode · @octo-patch](https://github.com/AtomFlow-AI/MoleCode/pull/3)
> Reason: Complete regional MiniMax transport support and make MiniMax-M3.1-Flash the default while retaining MiniMax-M3 and explicit saved model choices.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · Jozkah/flint · @Jozkah](https://github.com/Jozkah/flint/pull/1410)
> `web-app/src/constants/models.ts` listed MiniMax-M3 with `supportsImages: []` and had no MiniMax-M3.1-Flash, so Refresh / Add Model created M3 text-only.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · Levango7/MAOP · @Levango7](https://github.com/Levango7/MAOP/pull/263)
> - 批 6：models.yaml 官网列表价补录（MiniMax-M3 / qwen3.5-plus / gpt-5.4，每条钉 source URL + fetched_at）；agnes-2.5-pro 存量无来源价按官网刊例对齐 0.00045/0.0009；S-6 长未计价型号名 Badge 被省略号挤出 → flex 布局隔离
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · Lore-Hex/quill-cloud-proxy · @jperla](https://github.com/Lore-Hex/quill-cloud-proxy/pull/729)
> | `adaptive` effort or `thinking.type=adaptive` | kept for MiniMax M3; otherwise dropped (adaptive thinking falls back to reasoning_effort, or to the model default) |
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · NoFxAiOS/nofx · @octo-patch](https://github.com/NoFxAiOS/nofx/pull/1442)
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @SaladDay](https://github.com/NousResearch/hermes-agent/pull/141412)
> from the MiniMax team. This adds the new `MiniMax-M3.1-Flash` model to the `minimax`, `minimax-cn` and `minimax-oauth` model lists (ahead of `MiniMax-M3`, existing entries unchanged) so
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @PietjePuh](https://github.com/NousResearch/hermes-agent/pull/150270)
> - MiniMax-M3: `<|channel>thought\n…reasoning…<channel|>` (a tokenizer variant `<|channel|>thought` also occurs)
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @teknium1](https://github.com/NousResearch/hermes-agent/pull/151922)
> Before this change, MiniMax-M3's `<mm:think>` reasoning and the automatic "Compression refused" diagnostic both reached Telegram (and every other chat surface) as ordinary message text. Now
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @rsadler89](https://github.com/NousResearch/hermes-agent/pull/36873)
> /model minimax-m3:cloud
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @willhu-jpg](https://github.com/ROCm/aiter/pull/6742)
> fused shared experts, stage-two reduce variants, and published GLM-5/MiniMax-M3 native recipes retargeted to this geometry. The recipe sweep includes newer stage-two layouts and fused FP8
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · Scoheart/arca · @Scoheart](https://github.com/Scoheart/arca/pull/314)
> - Seeds: `MiniMax-M3`, `gemma-4-31B-it`, `gpt-oss-120b`. That was the entire public `GET /v1/models` body.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · agentgateway/agentgateway · @github-actions[bot]](https://github.com/agentgateway/agentgateway/pull/4347)
> json.providers.fireworks.models["accounts/fireworks/models/minimax-m3"] = {};
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · archestra-ai/archestra · @octo-patch](https://github.com/archestra-ai/archestra/pull/9283)
> selected M3 and older M2/highspeed models. This adds `MiniMax-M3.1-Flash` as the default and first recommendation, with `MiniMax-M3` retained as the other built-in choice.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · blairq/genesis-vllm-patches · @dependabot[bot]](https://github.com/blairq/genesis-vllm-patches/pull/7)
> XQA decode on SM12x (<a href="https://redirect.github.com/vllm-project/vllm/issues/60452">#60452</a>), and on MiniMax-M3 sparse attention on SM100 (<a href="https://redirect.github.com/vllm-project/vllm/issues/59300">#59300</a>); GLM-5.x defaults to an FP8 KV cache on SM100, halving
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · charmbracelet/catwalk · @adityabatra072](https://github.com/charmbracelet/catwalk/pull/658)
> GLM-5.3 (Fast) (`glm-5.3-fast`), Kimi K3 (`kimi-k3`), MiniMax M3 (`minimax-m3`), DeepSeek V4.1 Flash 1017 (`deepseek-v4.1-flash-1017`). Endpoint `https://inference.runanywhere.ai/v1`, key `$WALLY_API_KEY`, type `completions`.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · cheaper-inference/OmniRoute · @Adam-Dalloul](https://github.com/cheaper-inference/OmniRoute/pull/17209)
> it stays text-only. That also keeps the LEDGER-4 minimax-m3 vision check green.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · cosmicstack-labs/mercury-agent · @cloudgpuapp](https://github.com/cosmicstack-labs/mercury-agent/pull/264)
> `GET https://cloudgpu.app/v1/models` (2026-11-01): `deepseek-v4.1-flash`, `deepseek-v4-pro`, `kimi-k3`, `kimi-k2.7-code`, `glm-5.2`, `minimax-m3`, `qwen3-coder-480b`, `qwen3.5-397b`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · developerz-ai/ai-army · @developerz-ai[bot]](https://github.com/developerz-ai/ai-army/pull/93)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · developerz-ai/fast_rspec · @developerz-ai[bot]](https://github.com/developerz-ai/fast_rspec/pull/27)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · developerz-ai/mini · @developerz-ai[bot]](https://github.com/developerz-ai/mini/pull/36)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · developerz-ai/mini · @developerz-ai[bot]](https://github.com/developerz-ai/mini/pull/39)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · diegoami/borg-reborn-rs · @diegoami](https://github.com/diegoami/borg-reborn-rs/pull/66)
> - **Reviewer:** MiniMax-M3 (MiniMax), another model family, chosen through the quota service (rank 1).
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · diegoami/borg-reborn-rs · @diegoami](https://github.com/diegoami/borg-reborn-rs/pull/69)
> of #66 had been started by the session (MiniMax-M3, through the quota service), and is the reason for this rule. Docs only.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · elcattivo66/home-ops · @elcattivo66](https://github.com/elcattivo66/home-ops/pull/5840)
> LiteLLM routes `opencode-go/minimax-m3` as `anthropic/minimax-m3`. The Anthropic wire requires `max_tokens`, so LiteLLM injects its default **4096** when the client omits it. Hermes deliberately never sends
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · humumu130/cc-deck · @humumu130](https://github.com/humumu130/cc-deck/pull/94)
> 2. **G4 模型词表对齐 2.179.3 --help 实勘 17 键**：四幽灵键出列（glm-5.1/glm-5v-turbo/minimax-m3/**hy4-preview**——末者深测回执未列，本批 help 复勘多抓）+hy4free/hy4std 白名单误拒修复；漂移教训=CLI 升版动清单，custom: 前缀是逃生门
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3209)
> Kimi-K3 (with the DSPARK and DFlash drafter modules), MiniMax-M3, Muse Spark, Qwen3.5 and Qwen3.8-Flash-Next. So do the two per-commit Qwen3.5 PD and EPD unit tests
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/20)
> > Model `MiniMax-M3` on `router.ai.developerz.ai` via `native`, funded by `account_key`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/21)
> > Model `MiniMax-M3` on `router.ai.developerz.ai` via `native`, funded by `account_key`
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · sunholo-data/ailang-world · @sunholo-voight-kampff](https://github.com/sunholo-data/ailang-world/pull/523)
> `pi:openrouter/minimax/minimax-m3` (different model, different provider).
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · tuan3w/qwen-warp · @dependabot[bot]](https://github.com/tuan3w/qwen-warp/pull/11)
> XQA decode on SM12x (<a href="https://redirect.github.com/vllm-project/vllm/issues/60452">#60452</a>), and on MiniMax-M3 sparse attention on SM100 (<a href="https://redirect.github.com/vllm-project/vllm/issues/59300">#59300</a>); GLM-5.x defaults to an FP8 KV cache on SM100, halving
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @gordan12](https://github.com/vllm-project/vllm/pull/64320)
> centralized relay from #66177 (same opt-in as Qwen2, MiniMax-M3 and Qwen3-VL-MoE there); the Qwen3.5 counterpart is #61180.
- 2026-11-15T16:44:37.804Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @AndreasKaratzas](https://github.com/vllm-project/vllm/pull/66397)
> - **Other models:** MiniMax-M3 and Kimi-K3 use a similar `PeerBuffer` build-in-step pattern. They did not fail in CI and are out of scope here.
- 2026-11-15T16:46:23.763Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/d1079a679808657f62d0cb68e900b1b999186a65)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T16:46:23.763Z · [GitHub search · MiniMax · commits · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/commit/0019434826bc270d129ca37ce30cb9e6d76cc7c0)
> --language-model-only (MiniMax-M3 and Kimi-K3 runs state it true) was refused first. The key
- 2026-11-15T16:46:23.763Z · [GitHub search · MiniMax · commits · vllm-project/speculators · @wguo7](https://github.com/vllm-project/speculators/commit/988ed16b1f91315942049da3093d88dba70c45d9)
> fix(models): use the Gemma-style final norm for MiniMax-M3 verifiers (#1360)
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · ROCm/ATOM · @demonsan](https://github.com/ROCm/ATOM/pull/2822)
> for the neutral aiter `sp_qkv_fused` v1 op on MiniMax-M3 SP4 attention, gated at init time by `aiter.sp_qkv_covers(...)` over the deployment's local-M cell set, `status()["routed_ops"]`, and
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · Tencent/WeKnora · @henrikdukefoss](https://github.com/Tencent/WeKnora/pull/5312)
> at models (`deepseek/deepseek-v4.1-flash`, `deepseek/deepseek-v4-pro`, `z-ai/glm-5.3`, `z-ai/glm-5.3-flash`, `qwen/qwen3.8-flash`, `moonshotai/kimi-k3`, `minimax/minimax-m3`, `anthropic/claude-sonnet-5.5`, `openai/gpt-6.1-terra`, `google/gemini-3.8-flash`). Ids, context windows and image input come from the public `GET https://api.stdcmpt.com/v1/models`. No
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · agentforce314/clawcodex · @henrikdukefoss](https://github.com/agentforce314/clawcodex/pull/980)
> StandardCompute`, suggested models (`StandardCompute`, `deepseek/deepseek-v4.1-flash`, `deepseek/deepseek-v4-pro`, `moonshotai/kimi-k3`, `z-ai/glm-5.3`, `minimax/minimax-m3`, `qwen/qwen3.7-plus`, `anthropic/claude-sonnet-5.5`), `STANDARDCOMPUTE_API_KEY`, aliases `standard-compute` / `standard_compute`
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · assafelovic/gpt-researcher · @octo-patch](https://github.com/assafelovic/gpt-researcher/pull/1924)
> - Recommend `MiniMax-M3.1-Flash` first and `MiniMax-M3` second. Default omitted MiniMax models to Flash while preserving saved IDs, custom IDs, and the SDK's `model_name` alias.
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · bacluc-agent/litellm · @bacluc-agent](https://github.com/bacluc-agent/litellm/pull/10)
> kimi-k2.7-code, longcat-2.0, step-5-preview, beam-free, mimo-v2.6-flash, mimo-v2.6-pro, mimo-v2.5, mimo-v2.5-pro, minimax-m3.1-flash, minimax-m3, qwen3.8-max, qwen3.8-flash, qwen3.7-plus, deepseek-v4.1-flash, deepseek-v4-pro, hy4-preview, hy3`
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · docker/docker-agent · @github-actions[bot]](https://github.com/docker/docker-agent/pull/4829)
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · microsoft/Ontology-Playground · @octo-patch](https://github.com/microsoft/Ontology-Playground/pull/158)
> - Register an opt-in `minimax` adapter with `MiniMax-M3.1-Flash` as its default and `MiniMax-M3` as the other documented current model. Explicit server-side `MINIMAX_MODEL` overrides pass through
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @shiyang814-cpu](https://github.com/sgl-project/sglang/pull/36095)
> [Fix] minimax-m3: Support pipeline parallelism in PD disaggregation
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @vorapolsiloai](https://github.com/vllm-project/vllm/pull/64929)
> (wrong outputs on models with large activations, e.g. MiniMax-M3: MMMU 0.27–0.30). The kernel fix is ROCm/aiter#7258 (approved); once vLLM's AITER pin includes it, this switch
- 2026-11-15T17:00:47.877Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @vorapolsiloai](https://github.com/vllm-project/vllm/pull/65352)
> treat it as the embedding dtype (`image_embeds.type(self.visual.dtype)`), and MiniMax-M3 already declares it. A new attribute would be a second declaration that could drift. The helper
- 2026-11-15T17:02:14.066Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/fb3b969f11628e2ef750de683ee91f97bd97c2eb)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T17:02:14.066Z · [GitHub search · MiniMax · commits · byrdbryan5830/byr · @byrdbryan5830](https://github.com/byrdbryan5830/byr/commit/f4d7b9f60568dff3a641bb75bb8effba17bd2067)
> Model Optimizer实战教程：如何将MiniMax-M3模型量化为NVFP4格式
- 2026-11-15T17:02:14.066Z · [GitHub search · MiniMax · commits · cbc02009/k8s-home-ops · @cbc02009](https://github.com/cbc02009/k8s-home-ops/commit/83c97f47b240b3121d0ddbc4a09b84d966c43fa5)
> - reasoning-pool: Go kimi-k2.7-code + minimax-m3 + local rungs
- 2026-11-15T17:02:14.066Z · [GitHub search · MiniMax · commits · gildrb/eukhe · @gildrb](https://github.com/gildrb/eukhe/commit/628a485fa498245e2237b57309db96707eaa8074)
> models and opencode-go minimax-m3.
- 2026-11-15T17:02:14.066Z · [GitHub search · MiniMax · commits · kingkillery/oh-my-pk · @kingkillery](https://github.com/kingkillery/oh-my-pk/commit/76a4b3de27267adc46f9e83c8ac8dbd824b1d196)
> refresh moved the minimax-code and minimax-code-cn defaults to MiniMax-M3.1-Flash but left this test asserting the old MiniMax-M3 default.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · NVIDIA/Megatron-LM · @conver334](https://github.com/NVIDIA/Megatron-LM/pull/9023)
> [dev] lite: Add MiniMax-M3 model
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @ischencheng](https://github.com/NVIDIA/TensorRT-LLM/pull/21540)
> [None][feat] Marlin MoE: run SwigluBias; default MiniMax-M3 to Marlin on Hopper
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · ROCm/ATOM · @ningding01](https://github.com/ROCm/ATOM/pull/1551)
> MiniMax-M3 dense attention layers (0-2) route extend through the plain varlen path (_forward_extend_mha), which builds cu_seqlens_q from the FULL seq_lens while q holds only the
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @leonling-ll](https://github.com/ROCm/aiter/pull/3902)
> This PR aims to support MiniMax-M3 functionally work and accuracy correct on MI455.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · Travis-Gilbert/Theorem-UI · @Travis-Gilbert](https://github.com/Travis-Gilbert/Theorem-UI/pull/44)
> gate on the local e2e stack (real OpenRouter minimax/minimax-m3, signed-in `claude-e2e`), thread "CPU vs GPU":
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · afonsoft/OmniRouteNet · @devin-ai-integration[bot]](https://github.com/afonsoft/OmniRouteNet/pull/101)
> via tools/call + payloads, caps mapeadas do registry (minimax-m3→vision, GLM-5.2→tools+reasoning), audit blob migra uma vez sem dup.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · agntn/archives · @renovate[bot]](https://github.com/agntn/archives/pull/225)
> turns now show the pay-as-you-go equivalent cost, with MiniMax-M3.1-Flash-Preview estimated at the MiniMax-M3 rate since it has no published price ([#&#8203;13695](https://redirect.github.com/can1357/oh-my-pi/pull/13695) by [@&#8203;eggpeat](https://redirect.github.com/eggpeat))
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · anchapin/fluxion · @anchapin](https://github.com/anchapin/fluxion/pull/4597)
> Worker: MiniMax-M3 (flagship), 60/60 steps, ~4,301,954 in / ~84,200 out tokens.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · anchapin/planar-nexus · @anchapin](https://github.com/anchapin/planar-nexus/pull/2904)
> Worker: flagship tier, minimax/MiniMax-M3, 60/60 steps, ~2.36M in / 38.4k out tokens. Third dispatch of #2903 under MA-Mesh failover (prior balanced runs: failed:bad-diff, failed:worker-error); this
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/25)
> > Model `MiniMax-M3` on `router.ai.developerz.ai` via `native`, funded by `account_key`
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3205)
> refactor(models): run MiniMax-M3 on the shared decoder layer
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · openagents-org/openagents · @octo-patch](https://github.com/openagents-org/openagents/pull/812)
> gateway model refresh can reintroduce retired choices. Recommend MiniMax-M3.1-Flash first and MiniMax-M3 next for direct access, while retaining explicitly saved model IDs.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · osborn-studio/osbornalpha · @gilosbornhello](https://github.com/osborn-studio/osbornalpha/pull/206)
> Portable receipt checks and `git diff --check` passed. MiniMax-M3 scope audit returned PASS.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · oskarkocol/mission-control-contrib · @cursor[bot]](https://github.com/oskarkocol/mission-control-contrib/pull/11)
> Align MiniMax-M3 token pricing in `token-pricing.ts` and `models.ts` with Narev's public standard pay-as-you-go tier ($0.30 / $1.20 / $0.06 per MTok for input / output
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · reacher-z/Awesome-CUA · @reacher-z](https://github.com/reacher-z/Awesome-CUA/pull/739)
> for gpt-5.6, 108 for the other six), and `website_demo/MiniMax-M3/tasks/001` (`2381fce2`: traj.jsonl, 501 PNG step screenshots, result files, eval/runtime logs, checkpoint results and API usage). Counted
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @tnguyeng](https://github.com/sgl-project/sglang/pull/44249)
> Mean `sglang_bench_serve` on MI355X, SGLang TP4, MiniMax-M3 MXFP4, INT4 all-reduce. Change is pattern 3,4 relative to stride 3. Negative TTFT or TPOT is lower latency.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · smg-project/smg · @Moersity](https://github.com/smg-project/smg/pull/3471)
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · smg-project/smg · @hello-alexmcc](https://github.com/smg-project/smg/pull/3567)
> | MiniMax-M3 (MiniMax) | chunked raw body, no `stream_options`, two `root` messages unmerged, client stream without a usage chunk; counter `streamed/pure_forward` | fixed-length body, `include_usage:
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · theexperiencecompany/gaia · @aryanranderiya](https://github.com/theexperiencecompany/gaia/pull/1673)
> the OpenRouter model registry (`OPENROUTER_MODEL_TOOL_IMAGE_SUPPORT`). The dev menu's `minimax/minimax-m3` and `z-ai/glm-5.2` pins moved into it, and `DevModelOption.provider_pin` is gone.
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · trycua/cua · @octo-patch](https://github.com/trycua/cua/pull/1185)
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @gau-nernst](https://github.com/vllm-project/vllm/pull/60376)
> shares dispatch decisions with warmup. Ordered tables tune MiniMax-M3/M2 and Hunyuan-V3/V4 shapes; other shapes use a default table. Large-M execution requires `K % 64 ==
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @amd-dlimpus](https://github.com/vllm-project/vllm/pull/65194)
> Adds an MXFP4 mode to MiniMax-M3's fused FlyDSL norm / RoPE / KV insert on ROCm (gfx950). The same launch that writes NVFP4 pages can
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @MrGeva](https://github.com/vllm-project/vllm/pull/65691)
> logits of models that opt in with `SupportsFP32LogitsGather` (MiniMax-M3 / M3.1). This PR applies the same fused gather to the DSpark draft head's base logits
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · zachlatta/freeflow · @octo-patch](https://github.com/zachlatta/freeflow/pull/285)
- 2026-11-15T17:16:22.526Z · [GitHub search · MiniMax · pulls · zeroclaw-labs/zeroclaw · @octo-patch](https://github.com/zeroclaw-labs/zeroclaw/pull/11871)
- 2026-11-15T17:18:13.826Z · [GitHub search · MiniMax · commits · RBLN-SW/vllm-rbln · @rebel-jinsolkim](https://github.com/RBLN-SW/vllm-rbln/commit/bdefa5243625bff5833fb5e4d46653664877f631)
> fix(minimax-m3): read the reasoning state from the generation prompt only (#1428)
- 2026-11-15T17:18:13.826Z · [GitHub search · MiniMax · commits · moirai-internal/vllm · @hmellor](https://github.com/moirai-internal/vllm/commit/92f3ee5ae4abb86f2b175b13e623dab4a21b0986)
> [Model] Move MiniMax-M3 to `AutoWeightsLoader` (#65904)
- 2026-11-15T17:18:13.826Z · [GitHub search · MiniMax · commits · vllm-project/vllm · @hmellor](https://github.com/vllm-project/vllm/commit/92f3ee5ae4abb86f2b175b13e623dab4a21b0986)
> [Model] Move MiniMax-M3 to `AutoWeightsLoader` (#65904)
- 2026-11-15T17:18:13.826Z · [GitHub search · MiniMax · commits · zhao6300/vllm-opt · @hmellor](https://github.com/zhao6300/vllm-opt/commit/92f3ee5ae4abb86f2b175b13e623dab4a21b0986)
> [Model] Move MiniMax-M3 to `AutoWeightsLoader` (#65904)
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · Artem535/TFEngineCode · @Artem535](https://github.com/Artem535/TFEngineCode/pull/44)
> - **MiniMax-M3:** **-63.8% tokens**, **-81.9% cost** (0.04386 ₽ vs 0.24220 ₽), 3.6x faster.
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · GeminiLight/MindOS · @octo-patch](https://github.com/GeminiLight/MindOS/pull/61)
> sh endpoint](https://openrouter.ai/api/v1/models/minimax/minimax-m3.1-flash:free/endpoints), [OpenRouter M3 endpoint](https://openrouter.ai/api/v1/models/minimax/minimax-m3/endpoints).
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · Jonah-May-OSS/wyoming-whisper-trt · @dependabot[bot]](https://github.com/Jonah-May-OSS/wyoming-whisper-trt/pull/1123)
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · Jonah-May-OSS/wyoming-whisper-trt · @renovate[bot]](https://github.com/Jonah-May-OSS/wyoming-whisper-trt/pull/1125)
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · ai-dynamo/frontend-crates · @philei-tt](https://github.com/ai-dynamo/frontend-crates/pull/566)
> MiniMax-M3 / M3.1 omit a tool-call argument they mean as `null`. MiniMax's own chat template renders `None` arguments by omitting the field (its comment: "the
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · developerz-ai/claude-task-master · @developerz-ai[bot]](https://github.com/developerz-ai/claude-task-master/pull/184)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · logicplanes/kaeso · @dvnold](https://github.com/logicplanes/kaeso/pull/798)
> 50, `anthropic/claude-haiku-5-5` has its own prices above 100K, `minimax/MiniMax-M3` above 512K, and Qwen models above 32K, 128K and 256K. Until now Kaeso read none of
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · unslothai/llama.cpp · @danielhanchen](https://github.com/unslothai/llama.cpp/pull/320)
- 2026-11-15T17:31:57.934Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @Pb314314](https://github.com/vllm-project/vllm/pull/65227)
> | MiniMax-M3 | 1.086x | 1.119x | 2 / 48 (60 tokens +2.2%, 3996 tokens +2.1%) | 25 → 36 | 4.99 → 1.79 GiB
- 2026-11-15T17:33:24.211Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/559fb2b28fb2f384717d30a95ab25f9960d45ebd)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T17:33:24.211Z · [GitHub search · MiniMax · commits · Nano-Collective/.github · @will-lamerton](https://github.com/Nano-Collective/.github/commit/a25d4487881447f3d3ff5dfb7acd4792473b08b2)
> nc-review hardcoded --model minimax-m3 in its run step while the provider
- 2026-11-15T17:33:24.211Z · [GitHub search · MiniMax · commits · Orkeon/orkeon · @arion-orkeon](https://github.com/Orkeon/orkeon/commit/206b5622214a3642de4a9bbdd98bbe2765d3040e)
> OpenRouter and Mammouth default to gemini-3.8-flash, MiniMax to MiniMax-M3 (LLM-08)
- 2026-11-15T17:33:24.211Z · [GitHub search · MiniMax · commits · SemiAnalysisAI/InferenceX-app · @hbarclay](https://github.com/SemiAnalysisAI/InferenceX-app/commit/39253b9280d2dc8fbbc6ac58b47e5fe817378b1b)
> run it. Counting it at bf16 put B300 MiniMax-M3 long prefills at up to 116% of roofline. Linear recurrences stay bf16.
- 2026-11-15T17:33:24.211Z · [GitHub search · MiniMax · commits · lavabyte/deepai-gateway · @lavabyte](https://github.com/lavabyte/deepai-gateway/commit/a148082db7191845d60d2d475bcefccab1b28858)
> Checking it live: 22 models, `minimax-m3` at `2 Credits/request`, `gpt-5.3-flash` at
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · CraftJarvis/JARVIS-1 · @octo-patch](https://github.com/CraftJarvis/JARVIS-1/pull/13)
> - Default MiniMax requests to `MiniMax-M3.1-Flash`, with `MiniMax-M3` as the other current model. Preserve explicitly configured legacy/custom IDs and URLs.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @ScotterMonk](https://github.com/NousResearch/hermes-agent/pull/156191)
> "route": {"model": "MiniMax-M3", "provider": "minimax", "routed": true},
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · Owlbay/modelinfo-data · @Owlbay](https://github.com/Owlbay/modelinfo-data/pull/70)
> | minimax-m3 | prices[USD] input / output / cache_read | 0.3 / 1.2 / 0.06 → **0.6 / 2.4 / 0.12** | [Pricing Overview](https://platform.minimax.io/docs/pricing/overview)：Standard 档「MiniMax-M3
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @Boss2002n](https://github.com/ROCm/aiter/pull/7579)
> to #7079 on gfx1250 PA decode, tuned on MiniMax-M3 attention shapes (H16/KV1/D128). The main change is wider softmax steps. With a single query tile, each
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · Scoheart/arca · @Scoheart](https://github.com/Scoheart/arca/pull/270)
> official models overview: `@makers/hy3`, `@makers/hy3-preview`, `@makers/deepseek-v4-pro`, `@makers/deepseek-v4-flash`, `@makers/deepseek-v4.1-flash`, `@makers/minimax-m3`,
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · diet103/claude-code-infrastructure-showcase · @octo-patch](https://github.com/diet103/claude-code-infrastructure-showcase/pull/31)
> interface against the MiniMax Anthropic-compatible endpoint (default model MiniMax-M3, default base URL https://api.minimax.io/anthropic, China region https://api.minimaxi.com/anthropic)\n- Register MiniMax in provider-factory.ts: import, ProviderName type, PROVIDER_CREATORS map,
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · ggml-org/llama.cpp · @ServeurpersoCom](https://github.com/ggml-org/llama.cpp/pull/31851)
> keeps the computation in F32 (DeepSeek V4/V4.1 routers, MiniMax-M3 indexer), the request stays and gets a comment linking the reference line. Where it does not
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · henryopencode/neurix-harness · @henryopencode](https://github.com/henryopencode/neurix-harness/pull/7)
> 进程 25856 自动启动；观察期间未手动打开应用。关于窗口显示 1.0.3，原有会话的 2 轮 4 步及 MiniMax-M3 / Default 保留。安装记录和旧应用暂存目录自动清除。安装后 7,518 个文件条目逐项与官方 DMG 一致，ad-hoc 签名验证通过。
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/pull/51)
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · janpuc/home-ops · @janpuc](https://github.com/janpuc/home-ops/pull/1756)
> about 3M tokens per review. Its review fallback, MiniMax-M3, is on the same plan, so it did nothing when the plan ran out (31 Oct,
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · johannesjo/parallel-code · @octo-patch](https://github.com/johannesjo/parallel-code/pull/247)
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · likec4/likec4 · @octo-patch](https://github.com/likec4/likec4/pull/3436)
> Chat Completions factory and selects Flash by default. `MINIMAX_CHAT_MODEL=MiniMax-M3`, explicit legacy/custom model IDs, and existing global/CN region aliases remain supported.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · logicplanes/kaeso · @dvnold](https://github.com/logicplanes/kaeso/pull/782)
> `MINIMAX_API_KEY=… cargo xtask record anthropic_messages --provider minimax --model MiniMax-M3 deep-thinking tool-use-auto image-url-stream …`. The request fixtures for that are in `providers/minimax/requests/`.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · mati-ai-labs/agents-studio · @aditya-dev-10](https://github.com/mati-ai-labs/agents-studio/pull/42)
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · matt1398/claude-devtools · @octo-patch](https://github.com/matt1398/claude-devtools/pull/227)
> - Recognize MiniMax-M3 and MiniMax-M2.7 in the shared model parser.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · router-for-me/CLIProxyAPI · @EricLi404](https://github.com/router-for-me/CLIProxyAPI/pull/3794)
> Some large-context models (e.g. DeepSeek V4 Pro, MiniMax-M3 1M, Mimo V2.5 Pro) support up to 1,000,000 tokens of context, but their upstream `/v1/models` endpoint may
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @kevin-mii](https://github.com/sgl-project/sglang/pull/46693)
> --tp 4 --reasoning-parser minimax-m3 --tool-call-parser minimax-m3 --speculative-algorithm DSPARK
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @kevin-mii](https://github.com/sgl-project/sglang/pull/46694)
> - MiniMax-M3 and M3-VL loaders: the llm-compressor expert remap and the MXFP8 `weight_scale` -> `weight_scale_inv` alias run only on HIP.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · surendra2304/FRIDAY-Universe · @surendra2304](https://github.com/surendra2304/FRIDAY-Universe/pull/492)
> - `benchmarks/candidates/minimax.json` — MiniMax-M3 via https://api.minimax.io/v1 (route/model id from MiniMax's public API docs, read 2026-11-15). Live CLI run: **BLOCKED_NO_CREDENTIALS** (MINIMAX_API_KEY not provisioned), exit code 2.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @kyleliang-nv](https://github.com/vllm-project/vllm/pull/54604)
> - FULL graph profiling and end-to-end MiniMax-M3 startup/inference validation remain pending. CPU tests do not validate device-stream ordering or CUDA graph replay; no model-accuracy claim
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @yeighta](https://github.com/vllm-project/vllm/pull/65664)
> back to the default latent-MoE path, as the MiniMax-M3 fused MoE finalize already does. Multi-GPU NVIDIA Confidential Computing is one setup without multicast.
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · ysq111333/verl_jev · @dependabot[bot]](https://github.com/ysq111333/verl_jev/pull/1)
- 2026-11-15T17:48:33.197Z · [GitHub search · MiniMax · pulls · zendev-sh/goai · @cloudgpuapp](https://github.com/zendev-sh/goai/pull/210)
> IDs are bare names (`deepseek-v4.1-flash`, `deepseek-v4-pro`, `kimi-k3`, `glm-5.3-flash`, `minimax-m3`), checked against the public `GET https://cloudgpu.app/v1/models` catalog.
- 2026-11-15T17:50:07.737Z · [GitHub search · MiniMax · commits · Huiyicc/ik_llama.cpp_glm_ocr · @mb8565](https://github.com/Huiyicc/ik_llama.cpp_glm_ocr/commit/4c1987cad2303ecbaa53b8a3a3dac6f7672bff0c)
> SITU_GLU. A BF16 MiniMax-M3 GGUF aborts with the default -fug, and with
- 2026-11-15T17:50:07.737Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/89a338e30ad5c55a007657c3eef10e8ceb9153cf)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)
- 2026-11-15T17:50:07.737Z · [GitHub search · MiniMax · commits · PhilAnderson1/minicpm-pi5 · @mb8565](https://github.com/PhilAnderson1/minicpm-pi5/commit/4c1987cad2303ecbaa53b8a3a3dac6f7672bff0c)
> SITU_GLU. A BF16 MiniMax-M3 GGUF aborts with the default -fug, and with
- 2026-11-15T17:50:07.737Z · [GitHub search · MiniMax · commits · ikawrakow/ik_llama.cpp · @mb8565](https://github.com/ikawrakow/ik_llama.cpp/commit/4c1987cad2303ecbaa53b8a3a3dac6f7672bff0c)
> SITU_GLU. A BF16 MiniMax-M3 GGUF aborts with the default -fug, and with
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · AI-Team-26/Use-AI-for-coding · @alex-cyber-75](https://github.com/AI-Team-26/Use-AI-for-coding/pull/117)
> cannot capture `34 tok/s`. A model name like `Openrouter/minimax/minimax-m3` is expected to contain slashes.
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · MayankBansal12/wavmo-playground · @MayankBansal12](https://github.com/MayankBansal12/wavmo-playground/pull/27)
> model settings with one `OPENCODE_MODEL` setting, defaulting to `minimax-m3`. Planning, writing, editing, reviewing, and improving all use the configured model; planner-generated model names cannot override
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @am423](https://github.com/NousResearch/hermes-agent/pull/150180)
> - **`hermes_cli/models_catalog_static.py`**: adds `MiniMax-M3.1-Flash` to the curated floor. `MiniMax-M3` stays at `[0]`: `get_default_model_for_provider` returns `models[0]`, so a newer flagship must not silently take the non-interactive
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · Significant-Gravitas/AutoGPT · @octo-patch](https://github.com/Significant-Gravitas/AutoGPT/pull/13619)
> and its existing Chat Completions route, and add MiniMax-M3.1-Flash and MiniMax-M3 to the text model catalog. Flash has the documented 1,000,000-token context, 524,288-token output limit,
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · anchapin/portkit-v2 · @anchapin](https://github.com/anchapin/portkit-v2/pull/452)
> Worker: flagship tier, minimax/MiniMax-M3, 33,741 tokens out, 51/60 steps, grade 90. Two earlier balanced attempts at #283 failed empty-diff; this narrowed brief (failover-mandated flagship) is
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · awesome-dsh-plugin/awesome-dsh-plugin · @Wahwayhale](https://github.com/awesome-dsh-plugin/awesome-dsh-plugin/pull/12619)
> GLM-5.3-Flash / GLM-5.3 / DeepSeek-V4.1-Flash / DeepSeek-V4-Pro / MiniMax-M3.1-Flash / MiniMax-M3 / Kimi-K3.1-Pro / Kimi-K3). Each model shows its credit-cost multiplier, as the app does.
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/24)
> > Model `MiniMax-M3` on `api.minimax.io` via `native`, funded by `account_key`
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · grandchildrice/meden · @grandchildrice](https://github.com/grandchildrice/meden/pull/10)
> Record the completed same-model benign comparisons (minimax-m3, four workers, AgentDojo 0.1.35). Native/baseline: banking 44/48 vs 42/48 over three passes, Slack 18/21 vs 18/21, travel 14/20
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · jsboige/CoursIA · @jsboige](https://github.com/jsboige/CoursIA/pull/22882)
> : `/v1/models` y liste `[glm-5.2, glm-5.1, glm-4.7, qwen3.6-35b-a3b, MiniMax-M3, glm-5.3, qwen3.8-max, qwen3.7-max, qwen3.7-plus, qwen3.6-flash, deepseek-v4-pro, deepseek-v4-flash-0731, qwen3.8-flash, deepseek-v4.1-flash]` — pas de `swift-1.5-27b`. Remplacer le défaut
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · jsboige/CoursIA · @myia-ai-01](https://github.com/jsboige/CoursIA/pull/24540)
> côté claudish liste 14 modèles, dont `glm-5.3`, `deepseek-v4.1-flash`, `MiniMax-M3`, `qwen3.6-35b-a3b` et `qwen3.8-flash`. Les autres modèles de la matrice demandée (Swift, Codex luna/sol, Anthropic haiku/opus, Ornith-1.5-9B)
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · kwaroran/Risuai · @octo-patch](https://github.com/kwaroran/Risuai/pull/1524)
> Add MiniMax-M3.1-Flash first alongside MiniMax-M3 in the ordinary model picker. Keep M2.7 registry entries resolvable for saved selections without offering them as new choices. The
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · quoroom-ai/room · @octo-patch](https://github.com/quoroom-ai/room/pull/15)
> Reason: Add MiniMax API execution with MiniMax-M3.1-Flash as the default and MiniMax-M3 as the other ordinary model choice, while preserving explicitly saved model IDs and
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · smg-project/bellwether · @hello-alexmcc](https://github.com/smg-project/bellwether/pull/289)
> feat(corpus): MiniMax M3 size-rule references from the engine's own processors at the pinned release and at main
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @nbommi19](https://github.com/vllm-project/vllm/pull/59979)
> [Model][MiniMax-M3] Split-KV indexer prefill scoring and sync-free prefill metadata
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @hmellor](https://github.com/vllm-project/vllm/pull/61813)
> model dumps with the `transformers-upgrade` skill's scripts, for `MiniMaxAI/MiniMax-M3`, `MiniMaxAI/MiniMax-M3-MXFP8`, `nvidia/MiniMax-M3-NVFP4` and `amd/MiniMax-M3-MXFP4`: 0 diffs in parameters and module attributes, for both the main model
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @amd-dlimpus](https://github.com/vllm-project/vllm/pull/65196)
> The server log shows the fused launch running: `MiniMax-M3 FFN launch: rows (1..16)`, 57 sparse layers per step. Batches above 16 rows use the stock
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @akii96](https://github.com/vllm-project/vllm/pull/65898)
> Wires MiniMax-M3's sparse indexer to use MXFP4 index scoring on gfx942 (MI325X/MI308X/MI300X) via the aiter FP4 MQA kernels from ROCm/aiter#7017. The MXFP4 indexer uses about
- 2026-11-15T18:07:44.690Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @akii96](https://github.com/vllm-project/vllm/pull/65899)
> Adds MXFP4 main K/V cache support for MiniMax-M3 on gfx942 (MI325X/MI308X/MI300X). Uses the same 8704 B MXFP4 page layout as gfx950. MXFP4 KV holds 1.5
- 2026-11-15T18:08:11.448Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/85ccee5919f332e90b9740ff86e015ed75e46a70)
> model id(s) OR does not list: ['deepseek/deepseek-v4p1-flash', 'deepseek/deepseek-v4p1-flash-1017', 'minimax/minimax-m3', 'moonshotai/kimi-k3-fast', 'nvidia/nemotron-3-ultra-550b-a55b'] (+3 more)

## קישורים לראיות

- [GitHub search · MiniMax · pulls · AI-Team-26/Use-AI-for-coding · @alex-cyber-75](https://github.com/AI-Team-26/Use-AI-for-coding/pull/117)
- [GitHub search · MiniMax · pulls · AMD-AGI/Primus-Turbo · @botaohu001](https://github.com/AMD-AGI/Primus-Turbo/pull/627)
- [GitHub search · MiniMax · pulls · AMD-AGI/Primus · @lhzhang333](https://github.com/AMD-AGI/Primus/pull/1418)
- [GitHub search · MiniMax · pulls · Artem535/TFEngineCode · @Artem535](https://github.com/Artem535/TFEngineCode/pull/44)
- [GitHub search · MiniMax · pulls · AtomFlow-AI/MoleCode · @octo-patch](https://github.com/AtomFlow-AI/MoleCode/pull/2)
- [GitHub search · MiniMax · pulls · AtomFlow-AI/MoleCode · @octo-patch](https://github.com/AtomFlow-AI/MoleCode/pull/3)
- [GitHub search · MiniMax · pulls · Cmochance/codex-app-transfer · @octo-patch](https://github.com/Cmochance/codex-app-transfer/pull/614)
- [GitHub search · MiniMax · pulls · CraftJarvis/JARVIS-1 · @octo-patch](https://github.com/CraftJarvis/JARVIS-1/pull/13)
- [GitHub search · MiniMax · pulls · DennisLoska/skills · @DennisLoska](https://github.com/DennisLoska/skills/pull/13)
- [GitHub search · MiniMax · pulls · Eco-Sphere/kv-cache-calculator · @Gongdayao](https://github.com/Eco-Sphere/kv-cache-calculator/pull/14)
- [GitHub search · MiniMax · pulls · EvolvingLMMs-Lab/lmms-eval · @Luodian](https://github.com/EvolvingLMMs-Lab/lmms-eval/pull/1628)
- [GitHub search · MiniMax · pulls · FelipeGazapina/pinned · @FelipeGazapina](https://github.com/FelipeGazapina/pinned/pull/108)
- [GitHub search · MiniMax · pulls · GeminiLight/MindOS · @octo-patch](https://github.com/GeminiLight/MindOS/pull/61)
- [GitHub search · MiniMax · commits · Huiyicc/ik_llama.cpp_glm_ocr · @mb8565](https://github.com/Huiyicc/ik_llama.cpp_glm_ocr/commit/4c1987cad2303ecbaa53b8a3a3dac6f7672bff0c)
- [GitHub search · MiniMax · commits · IHUI-INF-AI/IHUI-AI · @unknown](https://github.com/IHUI-INF-AI/IHUI-AI/commit/f2fa36983ffc765c285553755f814b73bd9ddc37)
- [GitHub search · MiniMax · pulls · Inferact/llm-multimodal · @liangjuf](https://github.com/Inferact/llm-multimodal/pull/4)
- [GitHub search · MiniMax · pulls · Jonah-May-OSS/wyoming-whisper-trt · @dependabot[bot]](https://github.com/Jonah-May-OSS/wyoming-whisper-trt/pull/1123)
- [GitHub search · MiniMax · pulls · Jonah-May-OSS/wyoming-whisper-trt · @renovate[bot]](https://github.com/Jonah-May-OSS/wyoming-whisper-trt/pull/1125)
- [GitHub search · MiniMax · pulls · Jozkah/flint · @Jozkah](https://github.com/Jozkah/flint/pull/1410)
- [GitHub search · MiniMax · pulls · Kotlin/kotlin-jupyter-libraries · @renovate[bot]](https://github.com/Kotlin/kotlin-jupyter-libraries/pull/774)
- [GitHub search · MiniMax · pulls · LMCache/LMCache · @rebel-jinhwan](https://github.com/LMCache/LMCache/pull/6027)
- [GitHub search · MiniMax · pulls · Levango7/MAOP · @Levango7](https://github.com/Levango7/MAOP/pull/263)
- [GitHub search · MiniMax · pulls · Lore-Hex/quill-cloud-proxy · @jperla](https://github.com/Lore-Hex/quill-cloud-proxy/pull/729)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/1100b65a33c347e1f5b5c805f3859a510949eb5f)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/559fb2b28fb2f384717d30a95ab25f9960d45ebd)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/6e6cc9a0ac0f774cdd83713031afc1c1d444c688)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/859c8e8d58f06953e59453bcb53e0b74cae8f697)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/85ccee5919f332e90b9740ff86e015ed75e46a70)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/89a338e30ad5c55a007657c3eef10e8ceb9153cf)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/a172d71984aaafda89af8184b1191975c9352b5c)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/d1079a679808657f62d0cb68e900b1b999186a65)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/d8247ccdde55069887af433da8b6d0caa2f32989)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/e71210da62fdf0044b0b5ad16dd7fb703c26ca3e)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @unknown](https://github.com/Lore-Hex/quill-router/commit/fb3b969f11628e2ef750de683ee91f97bd97c2eb)
- [GitHub search · MiniMax · pulls · MandalorianBatman/austere-axi · @MandalorianBatman](https://github.com/MandalorianBatman/austere-axi/pull/10)
- [GitHub search · MiniMax · pulls · MayankBansal12/wavmo-playground · @MayankBansal12](https://github.com/MayankBansal12/wavmo-playground/pull/27)
- [GitHub · MiniMax-AI/minimax-code · code](https://github.com/MiniMax-AI/minimax-code/blob/3c21163c85a9684030cb147bfef48c6284fa3c46/packages/agent-modules/context-manager/src/settings.ts)
- [GitHub search · MiniMax · commits · Mutinho/agents-dashboard · @unknown](https://github.com/Mutinho/agents-dashboard/commit/82553191868d8e82a5a8382e490c5325829d6053)
- [GitHub search · MiniMax · pulls · NVIDIA/Megatron-LM · @conver334](https://github.com/NVIDIA/Megatron-LM/pull/9023)
- [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @peihu-nv](https://github.com/NVIDIA/TensorRT-LLM/pull/20629)
- [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @pranav-nvidia](https://github.com/NVIDIA/TensorRT-LLM/pull/20848)
- [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @ischencheng](https://github.com/NVIDIA/TensorRT-LLM/pull/21540)
- [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @WeiHaocheng](https://github.com/NVIDIA/TensorRT-LLM/pull/21712)
- [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @yijingl-nvidia](https://github.com/NVIDIA/TensorRT-LLM/pull/21869)
- [GitHub search · MiniMax · commits · Nano-Collective/.github · @will-lamerton](https://github.com/Nano-Collective/.github/commit/a25d4487881447f3d3ff5dfb7acd4792473b08b2)
- [GitHub search · MiniMax · pulls · NoFxAiOS/nofx · @octo-patch](https://github.com/NoFxAiOS/nofx/pull/1442)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @SaladDay](https://github.com/NousResearch/hermes-agent/pull/141412)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @am423](https://github.com/NousResearch/hermes-agent/pull/150180)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @PietjePuh](https://github.com/NousResearch/hermes-agent/pull/150270)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @teknium1](https://github.com/NousResearch/hermes-agent/pull/151922)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @mecworks](https://github.com/NousResearch/hermes-agent/pull/152431)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @ScotterMonk](https://github.com/NousResearch/hermes-agent/pull/156191)
- [GitHub search · MiniMax · pulls · NousResearch/hermes-agent · @rsadler89](https://github.com/NousResearch/hermes-agent/pull/36873)
- [GitHub search · MiniMax · pulls · OpenOSINT/OpenOSINT · @octo-patch](https://github.com/OpenOSINT/OpenOSINT/pull/40)
- [GitHub search · MiniMax · commits · Orkeon/orkeon · @arion-orkeon](https://github.com/Orkeon/orkeon/commit/206b5622214a3642de4a9bbdd98bbe2765d3040e)
- [GitHub search · MiniMax · pulls · Owlbay/modelinfo-data · @Owlbay](https://github.com/Owlbay/modelinfo-data/pull/70)
- [GitHub search · MiniMax · pulls · Pb314314/vllm · @Pb314314](https://github.com/Pb314314/vllm/pull/21)
- [GitHub search · MiniMax · commits · PhilAnderson1/minicpm-pi5 · @mb8565](https://github.com/PhilAnderson1/minicpm-pi5/commit/4c1987cad2303ecbaa53b8a3a3dac6f7672bff0c)
- [GitHub search · MiniMax · pulls · Plocr/dsh-commandcode · @Mou7s](https://github.com/Plocr/dsh-commandcode/pull/1)
- [GitHub search · MiniMax · pulls · QuantumNous/new-api · @anhaohighness](https://github.com/QuantumNous/new-api/pull/8281)
- [GitHub search · MiniMax · commits · RBLN-SW/vllm-rbln · @rebel-jinsolkim](https://github.com/RBLN-SW/vllm-rbln/commit/bdefa5243625bff5833fb5e4d46653664877f631)
- [GitHub search · MiniMax · pulls · ROCm/ATOM · @ningding01](https://github.com/ROCm/ATOM/pull/1551)
- [GitHub search · MiniMax · pulls · ROCm/ATOM · @rbrugaro-amd](https://github.com/ROCm/ATOM/pull/2582)
- [GitHub search · MiniMax · pulls · ROCm/ATOM · @demonsan](https://github.com/ROCm/ATOM/pull/2822)
- [GitHub search · MiniMax · pulls · ROCm/ATOM · @XiaobingSuper](https://github.com/ROCm/ATOM/pull/3052)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @leonling-ll](https://github.com/ROCm/aiter/pull/3902)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @willhu-jpg](https://github.com/ROCm/aiter/pull/6742)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @AndreasKaratzas](https://github.com/ROCm/aiter/pull/7053)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @lirui927](https://github.com/ROCm/aiter/pull/7549)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @Boss2002n](https://github.com/ROCm/aiter/pull/7579)
- [GitHub search · MiniMax · pulls · Scoheart/arca · @Scoheart](https://github.com/Scoheart/arca/pull/270)
- [GitHub search · MiniMax · pulls · Scoheart/arca · @Scoheart](https://github.com/Scoheart/arca/pull/294)
- [GitHub search · MiniMax · pulls · Scoheart/arca · @Scoheart](https://github.com/Scoheart/arca/pull/314)
- [GitHub search · MiniMax · commits · SemiAnalysisAI/InferenceX-app · @hbarclay](https://github.com/SemiAnalysisAI/InferenceX-app/commit/39253b9280d2dc8fbbc6ac58b47e5fe817378b1b)
- [GitHub search · MiniMax · pulls · Significant-Gravitas/AutoGPT · @octo-patch](https://github.com/Significant-Gravitas/AutoGPT/pull/13619)
- [GitHub search · MiniMax · pulls · Tencent/WeKnora · @henrikdukefoss](https://github.com/Tencent/WeKnora/pull/5312)
- [GitHub search · MiniMax · pulls · TencentQQGYLab/AppAgent · @octo-patch](https://github.com/TencentQQGYLab/AppAgent/pull/127)
- [GitHub search · MiniMax · pulls · Travis-Gilbert/Theorem-UI · @Travis-Gilbert](https://github.com/Travis-Gilbert/Theorem-UI/pull/44)
- [GitHub search · MiniMax · pulls · a-mansari/tarik · @a-mansari](https://github.com/a-mansari/tarik/pull/209)
- [GitHub search · MiniMax · pulls · afonsoft/OmniRouteNet · @devin-ai-integration[bot]](https://github.com/afonsoft/OmniRouteNet/pull/101)
- [GitHub search · MiniMax · pulls · agentforce314/clawcodex · @henrikdukefoss](https://github.com/agentforce314/clawcodex/pull/980)
- [GitHub search · MiniMax · pulls · agentgateway/agentgateway · @github-actions[bot]](https://github.com/agentgateway/agentgateway/pull/4347)
- [GitHub search · MiniMax · pulls · agntn/archives · @renovate[bot]](https://github.com/agntn/archives/pull/225)
- [GitHub search · MiniMax · pulls · ai-dynamo/dynamo · @furionw](https://github.com/ai-dynamo/dynamo/pull/17027)
- [GitHub search · MiniMax · pulls · ai-dynamo/frontend-crates · @philei-tt](https://github.com/ai-dynamo/frontend-crates/pull/566)
- [GitHub search · MiniMax · pulls · alibaba/open-code-review · @octo-patch](https://github.com/alibaba/open-code-review/pull/1784)
- [GitHub search · MiniMax · pulls · anchapin/fluxion · @anchapin](https://github.com/anchapin/fluxion/pull/4597)
- [GitHub search · MiniMax · pulls · anchapin/fluxion · @anchapin](https://github.com/anchapin/fluxion/pull/4651)
- [GitHub search · MiniMax · pulls · anchapin/fluxion · @anchapin](https://github.com/anchapin/fluxion/pull/4723)
- [GitHub search · MiniMax · pulls · anchapin/planar-nexus · @anchapin](https://github.com/anchapin/planar-nexus/pull/2884)
- [GitHub search · MiniMax · pulls · anchapin/planar-nexus · @anchapin](https://github.com/anchapin/planar-nexus/pull/2904)
- [GitHub search · MiniMax · pulls · anchapin/portkit-v2 · @anchapin](https://github.com/anchapin/portkit-v2/pull/452)
- [GitHub search · MiniMax · commits · angelstreet/mycodebox · @unknown](https://github.com/angelstreet/mycodebox/commit/850e1c87a2339923ebfca0aae97153ace7602d2f)
- [GitHub search · MiniMax · pulls · anymorph-ai/Claudable · @octo-patch](https://github.com/anymorph-ai/Claudable/pull/86)
- [GitHub search · MiniMax · pulls · archestra-ai/archestra · @octo-patch](https://github.com/archestra-ai/archestra/pull/9283)
- [GitHub search · MiniMax · pulls · arcships/zdr-monitor · @cunninghamcard-bit](https://github.com/arcships/zdr-monitor/pull/1166)
- [GitHub search · MiniMax · pulls · ashhart/TensorFold · @ashhart](https://github.com/ashhart/TensorFold/pull/1749)
- [GitHub search · MiniMax · pulls · assafelovic/gpt-researcher · @octo-patch](https://github.com/assafelovic/gpt-researcher/pull/1924)
- [GitHub search · MiniMax · pulls · awesome-dsh-plugin/awesome-dsh-plugin · @Wahwayhale](https://github.com/awesome-dsh-plugin/awesome-dsh-plugin/pull/12619)
- [GitHub search · MiniMax · pulls · awesome-dsh-plugin/awesome-dsh-plugin · @Wahwayhale](https://github.com/awesome-dsh-plugin/awesome-dsh-plugin/pull/12938)
- [GitHub search · MiniMax · pulls · b1tninja/continuous · @b1tninja](https://github.com/b1tninja/continuous/pull/513)
- [GitHub search · MiniMax · pulls · bacluc-agent/litellm · @bacluc-agent](https://github.com/bacluc-agent/litellm/pull/10)
- [GitHub search · MiniMax · pulls · batipang/CC8 · @batipang](https://github.com/batipang/CC8/pull/493)
- [GitHub search · MiniMax · pulls · blairq/genesis-vllm-patches · @dependabot[bot]](https://github.com/blairq/genesis-vllm-patches/pull/7)
- [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/54a45d9f4d853314b61407387674f484b03730c6)
- [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/6e8ec57325d12347e7644f60d3196974db97a6e8)
- [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/7f8bd83f8f4363133c7d274782359c49a805aa21)
- [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/d6eb81e4d4c2ea0f15ef32b08480b92df24c3795)
- [GitHub search · MiniMax · commits · bniklsd-ui/trading-bot-lizard-branch · @bniklsd-ui](https://github.com/bniklsd-ui/trading-bot-lizard-branch/commit/ec9335a877acfe5e8bf1a7aeef0384a9d2f7037f)
- [GitHub search · MiniMax · commits · byrdbryan5830/byr · @byrdbryan5830](https://github.com/byrdbryan5830/byr/commit/f4d7b9f60568dff3a641bb75bb8effba17bd2067)
- [GitHub search · MiniMax · pulls · can1357/oh-my-pi · @mikucat0309](https://github.com/can1357/oh-my-pi/pull/18101)
- [GitHub search · MiniMax · commits · cbc02009/k8s-home-ops · @cbc02009](https://github.com/cbc02009/k8s-home-ops/commit/83c97f47b240b3121d0ddbc4a09b84d966c43fa5)
- [GitHub search · MiniMax · pulls · charmbracelet/catwalk · @adityabatra072](https://github.com/charmbracelet/catwalk/pull/658)
- [GitHub search · MiniMax · pulls · cheaper-inference/OmniRoute · @Adam-Dalloul](https://github.com/cheaper-inference/OmniRoute/pull/17209)
- [GitHub search · MiniMax · pulls · cosmicstack-labs/mercury-agent · @cloudgpuapp](https://github.com/cosmicstack-labs/mercury-agent/pull/264)
- [GitHub search · MiniMax · pulls · developerz-ai/ai-army · @developerz-ai[bot]](https://github.com/developerz-ai/ai-army/pull/100)
- [GitHub search · MiniMax · pulls · developerz-ai/ai-army · @developerz-ai[bot]](https://github.com/developerz-ai/ai-army/pull/93)
- [GitHub search · MiniMax · pulls · developerz-ai/ai-gaming-agent-mcp · @developerz-ai[bot]](https://github.com/developerz-ai/ai-gaming-agent-mcp/pull/27)
- [GitHub search · MiniMax · pulls · developerz-ai/ai-gaming-agent-mcp · @developerz-ai[bot]](https://github.com/developerz-ai/ai-gaming-agent-mcp/pull/29)
- [GitHub search · MiniMax · pulls · developerz-ai/claude-task-master · @developerz-ai[bot]](https://github.com/developerz-ai/claude-task-master/pull/184)
- [GitHub search · MiniMax · pulls · developerz-ai/claude-task-master · @developerz-ai[bot]](https://github.com/developerz-ai/claude-task-master/pull/199)
- [GitHub search · MiniMax · pulls · developerz-ai/claude-task-master · @developerz-ai[bot]](https://github.com/developerz-ai/claude-task-master/pull/201)
- [GitHub search · MiniMax · pulls · developerz-ai/db-mcp-gateway · @developerz-ai[bot]](https://github.com/developerz-ai/db-mcp-gateway/pull/277)
- [GitHub search · MiniMax · pulls · developerz-ai/fast_rspec · @developerz-ai[bot]](https://github.com/developerz-ai/fast_rspec/pull/27)
- [GitHub search · MiniMax · pulls · developerz-ai/mini · @developerz-ai[bot]](https://github.com/developerz-ai/mini/pull/36)
- [GitHub search · MiniMax · pulls · developerz-ai/mini · @developerz-ai[bot]](https://github.com/developerz-ai/mini/pull/39)
- [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/24)
- [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/25)
- [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/35)
- [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/36)
- [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/37)
- [GitHub search · MiniMax · pulls · developerz-ai/nexus-engine · @developerz-ai[bot]](https://github.com/developerz-ai/nexus-engine/pull/40)
- [GitHub search · MiniMax · pulls · developerz-ai/open-mmorpg · @developerz-ai[bot]](https://github.com/developerz-ai/open-mmorpg/pull/115)
- [GitHub search · MiniMax · pulls · diegoami/borg-reborn-rs · @diegoami](https://github.com/diegoami/borg-reborn-rs/pull/66)
- [GitHub search · MiniMax · pulls · diegoami/borg-reborn-rs · @diegoami](https://github.com/diegoami/borg-reborn-rs/pull/69)
- [GitHub search · MiniMax · pulls · diegoami/malpaco · @diegoami](https://github.com/diegoami/malpaco/pull/474)
- [GitHub search · MiniMax · pulls · diet103/claude-code-infrastructure-showcase · @octo-patch](https://github.com/diet103/claude-code-infrastructure-showcase/pull/31)
- [GitHub search · MiniMax · pulls · dimagi/open-chat-studio · @claude[bot]](https://github.com/dimagi/open-chat-studio/pull/4989)
- [GitHub search · MiniMax · pulls · docker/docker-agent · @github-actions[bot]](https://github.com/docker/docker-agent/pull/4829)
- [GitHub search · MiniMax · pulls · dzianisv/hermes-agent · @dzianisv](https://github.com/dzianisv/hermes-agent/pull/40)
- [GitHub search · MiniMax · pulls · elcattivo66/home-ops · @elcattivo66](https://github.com/elcattivo66/home-ops/pull/5840)
- [GitHub search · MiniMax · pulls · enkryptai/helm-charts · @vvemulapalli11](https://github.com/enkryptai/helm-charts/pull/174)
- [GitHub search · MiniMax · commits · evorule/evo-agent · @EvoRuleLab](https://github.com/evorule/evo-agent/commit/9f31f3cd5fd1348283c61efcae54f1f02ed2465b)
- [GitHub search · MiniMax · pulls · flytiger-eco/sglang-for-sail · @plusbang](https://github.com/flytiger-eco/sglang-for-sail/pull/256)
- [GitHub search · MiniMax · pulls · ggml-org/llama.cpp · @ServeurpersoCom](https://github.com/ggml-org/llama.cpp/pull/31851)
- [GitHub search · MiniMax · pulls · gi-dellav/zerostack · @octo-patch](https://github.com/gi-dellav/zerostack/pull/320)
- [GitHub search · MiniMax · commits · gildrb/eukhe · @gildrb](https://github.com/gildrb/eukhe/commit/628a485fa498245e2237b57309db96707eaa8074)
- [GitHub search · MiniMax · commits · gildrb/eukhe · @gildrb](https://github.com/gildrb/eukhe/commit/d845ed27d0d3760982fac83de2845023336801ff)
- [GitHub search · MiniMax · pulls · grandchildrice/meden · @grandchildrice](https://github.com/grandchildrice/meden/pull/10)
- [GitHub search · MiniMax · pulls · gudlyf/blynky · @gudlyf](https://github.com/gudlyf/blynky/pull/32)
- [GitHub search · MiniMax · pulls · hao-aaron/vllm · @hao-aaron](https://github.com/hao-aaron/vllm/pull/15)
- [GitHub search · MiniMax · pulls · henryopencode/neurix-harness · @henryopencode](https://github.com/henryopencode/neurix-harness/pull/7)
- [GitHub search · MiniMax · pulls · humumu130/cc-deck · @humumu130](https://github.com/humumu130/cc-deck/pull/94)
- [GitHub search · MiniMax · commits · ikawrakow/ik_llama.cpp · @mb8565](https://github.com/ikawrakow/ik_llama.cpp/commit/4c1987cad2303ecbaa53b8a3a3dac6f7672bff0c)
- [GitHub search · MiniMax · pulls · ikawrakow/ik_llama.cpp · @mb8565](https://github.com/ikawrakow/ik_llama.cpp/pull/2885)
- [GitHub search · MiniMax · pulls · im-dongseon/wikihub · @im-dongseon](https://github.com/im-dongseon/wikihub/pull/224)
- [GitHub search · MiniMax · commits · inference-sim/blis-substrate · @sriumcp](https://github.com/inference-sim/blis-substrate/commit/0ae394c26d3059536dd56a6fa18c2fed919a55a0)
- [GitHub search · MiniMax · commits · inference-sim/blis-substrate · @sriumcp](https://github.com/inference-sim/blis-substrate/commit/83a95d7a483ccd1b7c07548df484653f545dd44b)
- [GitHub search · MiniMax · commits · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/commit/0019434826bc270d129ca37ce30cb9e6d76cc7c0)
- [GitHub search · MiniMax · pulls · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/pull/51)
- [GitHub search · MiniMax · pulls · janpuc/home-ops · @janpuc](https://github.com/janpuc/home-ops/pull/1756)
- [GitHub search · MiniMax · pulls · johannesjo/parallel-code · @octo-patch](https://github.com/johannesjo/parallel-code/pull/247)
- [GitHub search · MiniMax · pulls · jsavoy93/trading-bot · @jsavoy93](https://github.com/jsavoy93/trading-bot/pull/141)
- [GitHub search · MiniMax · pulls · jsboige/CoursIA · @jsboige](https://github.com/jsboige/CoursIA/pull/22882)
- [GitHub search · MiniMax · pulls · jsboige/CoursIA · @myia-ai-01](https://github.com/jsboige/CoursIA/pull/24540)
- [GitHub search · MiniMax · pulls · jundot/omlx · @jundot](https://github.com/jundot/omlx/pull/5083)
- [GitHub search · MiniMax · commits · kingkillery/oh-my-pk · @kingkillery](https://github.com/kingkillery/oh-my-pk/commit/76a4b3de27267adc46f9e83c8ac8dbd824b1d196)
- [GitHub search · MiniMax · pulls · kwaroran/Risuai · @octo-patch](https://github.com/kwaroran/Risuai/pull/1524)
- [GitHub search · MiniMax · pulls · lamtuanvu/pacer · @lamtuanvu](https://github.com/lamtuanvu/pacer/pull/833)
- [GitHub search · MiniMax · commits · lavabyte/deepai-gateway · @lavabyte](https://github.com/lavabyte/deepai-gateway/commit/a148082db7191845d60d2d475bcefccab1b28858)
- [GitHub search · MiniMax · pulls · leehack/llamadart · @leehack](https://github.com/leehack/llamadart/pull/1894)
- [GitHub search · MiniMax · pulls · leehack/llamadart · @leehack](https://github.com/leehack/llamadart/pull/1910)
- [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3205)
- [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3209)
- [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3239)
- [GitHub search · MiniMax · pulls · lightseekorg/tokenspeed · @nperrin-fr](https://github.com/lightseekorg/tokenspeed/pull/3241)
- [GitHub search · MiniMax · pulls · likec4/likec4 · @octo-patch](https://github.com/likec4/likec4/pull/3436)
- [GitHub search · MiniMax · pulls · logicplanes/kaeso · @dvnold](https://github.com/logicplanes/kaeso/pull/782)
- [GitHub search · MiniMax · pulls · logicplanes/kaeso · @dvnold](https://github.com/logicplanes/kaeso/pull/798)
- [GitHub search · MiniMax · pulls · mati-ai-labs/agents-studio · @aditya-dev-10](https://github.com/mati-ai-labs/agents-studio/pull/42)
- [GitHub search · MiniMax · pulls · matt1398/claude-devtools · @octo-patch](https://github.com/matt1398/claude-devtools/pull/227)
- [GitHub search · MiniMax · pulls · microsoft/Ontology-Playground · @octo-patch](https://github.com/microsoft/Ontology-Playground/pull/158)
- [GitHub search · MiniMax · commits · moirai-internal/vllm · @hmellor](https://github.com/moirai-internal/vllm/commit/92f3ee5ae4abb86f2b175b13e623dab4a21b0986)
- [GitHub search · MiniMax · pulls · nexu-io/open-design · @orionp13](https://github.com/nexu-io/open-design/pull/8196)
- [GitHub search · MiniMax · pulls · onyx-dot-app/onyx · @octo-patch](https://github.com/onyx-dot-app/onyx/pull/12918)
- [GitHub search · MiniMax · pulls · openagents-org/openagents · @octo-patch](https://github.com/openagents-org/openagents/pull/812)
- [GitHub search · MiniMax · pulls · osborn-studio/osbornalpha · @gilosbornhello](https://github.com/osborn-studio/osbornalpha/pull/206)
- [GitHub search · MiniMax · pulls · oskarkocol/mission-control-contrib · @cursor[bot]](https://github.com/oskarkocol/mission-control-contrib/pull/11)
- [GitHub search · MiniMax · pulls · quoroom-ai/room · @octo-patch](https://github.com/quoroom-ai/room/pull/15)
- [GitHub search · MiniMax · pulls · reacher-z/Awesome-CUA · @reacher-z](https://github.com/reacher-z/Awesome-CUA/pull/739)
- [GitHub search · MiniMax · pulls · router-for-me/CLIProxyAPI · @EricLi404](https://github.com/router-for-me/CLIProxyAPI/pull/3794)
- [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/20)
- [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/21)
- [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/24)
- [GitHub search · MiniMax · pulls · sebyx07/ruby2html · @developerz-ai[bot]](https://github.com/sebyx07/ruby2html/pull/25)
- [GitHub search · MiniMax · pulls · sergiodelgadotech/kotlin-saas-template · @renovate[bot]](https://github.com/sergiodelgadotech/kotlin-saas-template/pull/181)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @shiyang814-cpu](https://github.com/sgl-project/sglang/pull/36095)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @waqahmed-amd-fi](https://github.com/sgl-project/sglang/pull/39994)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @tnguyeng](https://github.com/sgl-project/sglang/pull/44249)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @kevin-mii](https://github.com/sgl-project/sglang/pull/46693)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @kevin-mii](https://github.com/sgl-project/sglang/pull/46694)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @michaelzhang-ai](https://github.com/sgl-project/sglang/pull/46944)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @hnyls2002](https://github.com/sgl-project/sglang/pull/48776)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @Kangyan-Zhou](https://github.com/sgl-project/sglang/pull/49392)
- [GitHub search · MiniMax · pulls · sidalit/freetoken-snap · @renovate[bot]](https://github.com/sidalit/freetoken-snap/pull/1)
- [GitHub search · MiniMax · pulls · smg-project/bellwether · @hello-alexmcc](https://github.com/smg-project/bellwether/pull/289)
- [GitHub search · MiniMax · pulls · smg-project/smg · @Moersity](https://github.com/smg-project/smg/pull/3471)
- [GitHub search · MiniMax · pulls · smg-project/smg · @hello-alexmcc](https://github.com/smg-project/smg/pull/3567)
- [GitHub search · MiniMax · pulls · smg-project/smg · @hello-alexmcc](https://github.com/smg-project/smg/pull/5243)
- [GitHub search · MiniMax · pulls · sunholo-data/ailang-world · @sunholo-voight-kampff](https://github.com/sunholo-data/ailang-world/pull/523)
- [GitHub search · MiniMax · pulls · surendra2304/FRIDAY-Universe · @surendra2304](https://github.com/surendra2304/FRIDAY-Universe/pull/492)
- [GitHub search · MiniMax · pulls · theexperiencecompany/gaia · @aryanranderiya](https://github.com/theexperiencecompany/gaia/pull/1673)
- [GitHub search · MiniMax · pulls · trycua/cua · @octo-patch](https://github.com/trycua/cua/pull/1185)
- [GitHub search · MiniMax · pulls · tuan3w/qwen-warp · @dependabot[bot]](https://github.com/tuan3w/qwen-warp/pull/11)
- [GitHub search · MiniMax · pulls · unslothai/llama.cpp · @danielhanchen](https://github.com/unslothai/llama.cpp/pull/320)
- [GitHub search · MiniMax · commits · use-agent-os/agent-os · @andreapn](https://github.com/use-agent-os/agent-os/commit/4c9ba83f609465703b30f3547ac526663ae88bb8)
- [GitHub search · MiniMax · pulls · vercel-labs/fx · @fazxes](https://github.com/vercel-labs/fx/pull/1658)
- [GitHub search · MiniMax · pulls · vllm-project/semantic-router · @octo-patch](https://github.com/vllm-project/semantic-router/pull/5108)
- [GitHub search · MiniMax · commits · vllm-project/speculators · @wguo7](https://github.com/vllm-project/speculators/commit/988ed16b1f91315942049da3093d88dba70c45d9)
- [GitHub search · MiniMax · commits · vllm-project/vllm-ascend · @FORFuture37](https://github.com/vllm-project/vllm-ascend/commit/514a647be60f6044bb542f7be2529619559b1a15)
- [GitHub search · MiniMax · commits · vllm-project/vllm-ascend · @dianWang325](https://github.com/vllm-project/vllm-ascend/commit/8767c6c401e60e1f9c8dc00b8821955192ce5737)
- [GitHub search · MiniMax · pulls · vllm-project/vllm-ascend · @yingyingzizi](https://github.com/vllm-project/vllm-ascend/pull/18428)
- [GitHub search · MiniMax · pulls · vllm-project/vllm-ascend · @ella1107](https://github.com/vllm-project/vllm-ascend/pull/21460)
- [GitHub search · MiniMax · pulls · vllm-project/vllm-gaudi · @pawel-olejniczak](https://github.com/vllm-project/vllm-gaudi/pull/1965)
- [GitHub search · MiniMax · commits · vllm-project/vllm · @hmellor](https://github.com/vllm-project/vllm/commit/92f3ee5ae4abb86f2b175b13e623dab4a21b0986)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @kyleliang-nv](https://github.com/vllm-project/vllm/pull/54604)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @nbommi19](https://github.com/vllm-project/vllm/pull/59979)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @gau-nernst](https://github.com/vllm-project/vllm/pull/60376)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @hmellor](https://github.com/vllm-project/vllm/pull/61813)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @AndreasKaratzas](https://github.com/vllm-project/vllm/pull/63486)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @gordan12](https://github.com/vllm-project/vllm/pull/64320)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @vorapolsiloai](https://github.com/vllm-project/vllm/pull/64929)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @amd-dlimpus](https://github.com/vllm-project/vllm/pull/65194)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @amd-dlimpus](https://github.com/vllm-project/vllm/pull/65196)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @zixi-qi](https://github.com/vllm-project/vllm/pull/65217)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @Pb314314](https://github.com/vllm-project/vllm/pull/65227)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @vorapolsiloai](https://github.com/vllm-project/vllm/pull/65311)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @vorapolsiloai](https://github.com/vllm-project/vllm/pull/65352)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @yeighta](https://github.com/vllm-project/vllm/pull/65664)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @MrGeva](https://github.com/vllm-project/vllm/pull/65691)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @akii96](https://github.com/vllm-project/vllm/pull/65898)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @akii96](https://github.com/vllm-project/vllm/pull/65899)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @Pb314314](https://github.com/vllm-project/vllm/pull/65972)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @esmeetu](https://github.com/vllm-project/vllm/pull/66203)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @AndreasKaratzas](https://github.com/vllm-project/vllm/pull/66370)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @AndreasKaratzas](https://github.com/vllm-project/vllm/pull/66397)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @tomasruizt](https://github.com/vllm-project/vllm/pull/66521)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @hao-aaron](https://github.com/vllm-project/vllm/pull/66640)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @ZhengGong-amd](https://github.com/vllm-project/vllm/pull/66695)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @ZhengGong-amd](https://github.com/vllm-project/vllm/pull/66706)
- [GitHub search · MiniMax · pulls · xorbitsai/inference · @David-Wu1119](https://github.com/xorbitsai/inference/pull/6051)
- [GitHub search · MiniMax · pulls · ysq111333/verl_jev · @dependabot[bot]](https://github.com/ysq111333/verl_jev/pull/1)
- [GitHub search · MiniMax · pulls · zachlatta/freeflow · @octo-patch](https://github.com/zachlatta/freeflow/pull/285)
- [GitHub search · MiniMax · pulls · zendev-sh/goai · @cloudgpuapp](https://github.com/zendev-sh/goai/pull/210)
- [GitHub search · MiniMax · pulls · zeroclaw-labs/zeroclaw · @octo-patch](https://github.com/zeroclaw-labs/zeroclaw/pull/11871)
- [GitHub search · MiniMax · commits · zhao6300/vllm-opt · @hmellor](https://github.com/zhao6300/vllm-opt/commit/92f3ee5ae4abb86f2b175b13e623dab4a21b0986)
- [review](https://huggingface.co/MiniMaxAI/MiniMax-M3)
- [hf-author-artifacts:MiniMaxAI](https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3/paths-info/d95e30cb912fe3021bb9c2ba66a8ce60b6c937a9)
- [models-dev](https://models.dev/#minimax-m3)
- [opencode-go](https://opencode.ai/inference/go/v1/models)
- [minimax-model-index](https://platform.minimax.io/docs/api-reference/text-anthropic-api)

## איתותים