# MiniMax M3

Ang MiniMax M3 ay isang modelo mula sa MiniMax na may bukas na weights, na inilabas noong 2026-06-01. Ang context window ay 1,000,000 token.

- Tagabuo: MiniMax
- Id: minimaxai/minimax-m3
- Ayon sa mga ulat: 2026-07-14T11:26:12.000Z
- Inilabas: 2026-06-01T00:00:00.000Z
- Pagkatukoy: 2026-09-17T00:52:24.374Z
  - [link sa ebidensiya](https://models.dev/#minimax-m3)
- Katayuan: Inilabas
- Bersyon ng mga timbang: 2026-06-12
  - [link sa ebidensiya](https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3/paths-info/d95e30cb912fe3021bb9c2ba66a8ce60b6c937a9)
- Kanonikal: https://alexgetman.com/fil/models/minimax/minimax-m3-677/

## Mga tanong

### Gaano kalaki ang context window ng MiniMax M3?

Ang context window ay 1,000,000 token.

### Magagamit ba ng publiko ang weights ng MiniMax M3?

Oo. Inilalathala ng MiniMax ang weights.

### Kailan inilabas ang MiniMax M3?

Noong 2026-06-01.

## Takdang panahon

- 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);
> }
> 
Hindi pinatutunayan ng pagbanggit na ito sa code ang isang ginagamit na identifier ng modelo.
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · MrGeva/vllm · @MrGeva](https://github.com/MrGeva/vllm/pull/3)
> 320 questions, concurrency 32, same node, together with MiniMax-M3 model changes):
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · cbc02009/container-images · @mahoro-ai](https://github.com/cbc02009/container-images/pull/2)
> `/v1/chat/completions` request to a messages-protocol Go model (e.g. `opencode_go/minimax-m3`, `qwen3.8-max`, `claude-haiku-5-5`) dies mid-stream with `TypeError: 'NoneType' object is not an iterator` → `APIConnectionError` 500 →
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6140)
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6141)
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6142)
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · maximhq/bifrost · @octo-patch](https://github.com/maximhq/bifrost/pull/5184)
> API. New text examples and integration defaults select `MiniMax-M3.1-Flash`, with `MiniMax-M3` as the other current choice. Default management model suggestions put those models first and
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · maximhq/bifrost · @jimaldon](https://github.com/maximhq/bifrost/pull/7124)
> model catalog datasheet already carries `tencent/deepseek-v4-pro`, `tencent/deepseek-v4-flash` and `tencent/minimax-m3`; live-only models are listed with blank pricing and can be priced via overrides.
- 2026-11-13T18:36:02.583Z · [GitHub search · MiniMax · pulls · overnightworks/speech · @dependabot[bot]](https://github.com/overnightworks/speech/pull/354)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · EvolvingLMMs-Lab/lmms-eval · @sahirema](https://github.com/EvolvingLMMs-Lab/lmms-eval/pull/1571)
> sample by frame rate ignore it, e.g. vLLM's MiniMax-M3 and Qwen3-VL loaders. On those, `max_frames_num` has no effect: long clips are sampled at the loader's
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · NVIDIA-NeMo/Megatron-Bridge · @pruprakash](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/6839)
> seeing `TEDotProductAttention`; GREEN 4/4, full file 86 passed, Gemma4/MiniMax-M3/Kimi-K2.5-VL/GPT-OSS unit tests 100 passed. E2E: gpt-oss-20b card inference (TP2 PP2 EP2, `--attention-backend local`) on this branch
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @ziyixiong-nv](https://github.com/NVIDIA/TensorRT-LLM/pull/21438)
> sampler, speculative and QSA tests; the MXFP8 and MiniMax-M3 dispatch tests; GSM8K with DSpark (TP4);
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · Pb314314/vllm · @Pb314314](https://github.com/Pb314314/vllm/pull/19)
> use sparse MLA (dense MLA, DeepSeek-V4 / V4.1, MiniMax-M3, non-MLA models) are not affected; the reason for each is in "Per-model check".
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · Pinvou/pinvou-agent · @asto18089](https://github.com/Pinvou/pinvou-agent/pull/841)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · Portkey-AI/models · @polease](https://github.com/Portkey-AI/models/pull/1032)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · Quarkgluonmixture/quarkspace · @github-actions[bot]](https://github.com/Quarkgluonmixture/quarkspace/pull/130)
> - `intelligence` — minimax-m3 (default) intelligence: catalog 45.4, archive 18.9
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · ROCm/ATOM · @rbrugaro-amd](https://github.com/ROCm/ATOM/pull/2701)
> does for the target (vLLM already did this). MiniMax-M3's DSpark draft now validates this config instead of building its own, and declares `draft_skips_online_quant` so its
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · ROCm/ATOM · @yhl-amd](https://github.com/ROCm/ATOM/pull/3194)
> under `multi`, which is the prefill-side setup for MiniMax-M3 PD with UMBP:
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @RElbers](https://github.com/ROCm/aiter/pull/7218)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · ROCm/aiter · @zhiding512](https://github.com/ROCm/aiter/pull/7619)
> On gfx1250, every MiniMax-M3 sparse attention call (decode, MTP verify, prefill) runs FlyDSL `pa_decode` (`pa_fwd_asm` does not support gfx1250), and dense layers use it with
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · anomalyco/models.dev · @rjuez00](https://github.com/anomalyco/models.dev/pull/13443)
> - `MiniMax-M3` is no longer listed and a request using that ID returns `model=MiniMax-M3.1-Flash`, so the old entry is removed rather than kept as an
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · ashhart/TensorFold · @ashhart](https://github.com/ashhart/TensorFold/pull/3123)
> CUDA decode attention (`window_attention`, shared by Qwen3.5/3.8, Llama, MiniMax-M3, Spark 2.5, Step5 and LFM): `window_rows` merged every chunk's partial for every row, 128 chunks at
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · boelabs/ai-gateway · @jhisdev](https://github.com/boelabs/ai-gateway/pull/22)
> - `bun run test` fails: `vercel.test.ts` pins `vercel/minimax/minimax-m3` reasoning as `fixed ["high"]`; the regenerated snapshot now says `openai_effort ["none","high"]`. Tests may not be edited by
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · ghuntley/how-to-build-a-coding-agent · @octo-patch](https://github.com/ghuntley/how-to-build-a-coding-agent/pull/20)
> Anthropic-compatible defaults to the standalone workshop agents, with MiniMax-M3.1-Flash as the default and MiniMax-M3 as the alternative.
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · goonercbot/hermes-agent · @goonercbot](https://github.com/goonercbot/hermes-agent/pull/39)
> - Real configured MiniMax-M3 request through corrected Hermes adapter returned exactly `RECOVERY_OK`, end_turn, original request unchanged.
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/pull/134)
> the figures on the H200 rows and the MiniMax-M3 points are development-split, in-sample numbers.
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/pull/99)
> (`vllm/model_executor/models/config.py:43-55`) for every graph that is not DeepSeek-V4; MiniMax-M3's pool at the same memory is 72645 blocks, not 72893, because the catalog graph now holds
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · inference-sim/inference-sim · @sriumcp](https://github.com/inference-sim/inference-sim/pull/2104)
> the substrate's layer 8 and 9 price changes (MiniMax-M3 routers replicated in fp32, MXFP8 scale bytes, dsv4-nvfp4 MTP experts at MXFP4, a shared lm_head for
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6143)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · jundot/omlx · @begcdn](https://github.com/jundot/omlx/pull/5239)
> On M2 Ultra (`applegpu_g14d`) the MiniMax-M3 sparse prefill path crashes once a prompt crosses 4096 tokens: `msa_prefill_supported` gates only on shape (`D=128`, `block_size=128`, `qhead_per_kv=16`), so
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · openJiuwen-ai/jiuwenswarm · @octo-patch](https://github.com/openJiuwen-ai/jiuwenswarm/pull/2117)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · sgl-project/sglang · @alphabetc1](https://github.com/sgl-project/sglang/pull/50334)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · tenstorrent/tt-metal · @mdragulaTT](https://github.com/tenstorrent/tt-metal/pull/64261)
> The MiniMax-M3 vision encoder on one Blackhole chip. It turns one image, as the media module's `prepare` resizes it, into its vision tokens: bf16 `[n_tokens,
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · tenstorrent/tt-metal · @mdragulaTT](https://github.com/tenstorrent/tt-metal/pull/64459)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · thequanhcm/quanpt1 · @thequanhcm](https://github.com/thequanhcm/quanpt1/pull/28)
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · viztor/dsh-opencode-patch · @viztor](https://github.com/viztor/dsh-opencode-patch/pull/113)
> | `opencode-go/minimax-m3` | info | `[KNOWN_DEFECT]` toggle-only model offered Minimal…Max | Off / On (D1) |
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @wolfcomos](https://github.com/vllm-project/vllm/pull/62098)
> Add `minimax_m3_msa_decode_fp16_dot`, an opt-in optimization for MiniMax-M3 Triton sparse decode on SM100-family GPUs. It uses FP16 dot operands with FP32 accumulation and selects splits targeting
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @afriedri](https://github.com/vllm-project/vllm/pull/64089)
> short prompts at high concurrency is higher on MiniMax-M3 (TP8: up to +21% at 1024 tokens, conc 64; TP4: +14% at 1024
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @MrGeva](https://github.com/vllm-project/vllm/pull/65433)
> [Perf][MiniMax-M3] Raise the fused MoE finalize + all-reduce cap from 32 to 128 tokens
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @MrGeva](https://github.com/vllm-project/vllm/pull/65434)
> [Perf][MiniMax-M3] Two-shot fused MoE finalize + all-reduce + norm + MXFP8 up to 512 tokens
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @cjackal](https://github.com/vllm-project/vllm/pull/65453)
> - Pixtral, Cohere Compass and MiniMax-M3 also call `get_cu_seqlens_padding_logics`, so every encoder-CG model whose `cu_seqlens` comes from `maybe_recompute_cu_seqlens` takes its padding from one place. Without
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @Pb314314](https://github.com/vllm-project/vllm/pull/67338)
> use sparse MLA (dense MLA, DeepSeek-V4 / V4.1, MiniMax-M3, non-MLA models) are not affected; the reason for each is in "Per-model check".
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · vllm-project/vllm · @Vincent-PRO-AI](https://github.com/vllm-project/vllm/pull/67518)
> [Model][MiniMax-M3] SM120 Blackwell NVFP4 KV cache Triton reader
- 2026-11-13T19:04:38.386Z · [GitHub search · MiniMax · pulls · xiaoY233/Chat2API · @octo-patch](https://github.com/xiaoY233/Chat2API/pull/184)
> feat(minimax): add MiniMax-M3 model entry
- 2026-11-13T19:05:09.193Z · [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @jperla](https://github.com/Lore-Hex/quill-router/commit/664e5541dd2f40274ca93378ac783395e1c3576c)
- 2026-11-13T19:05:09.193Z · [GitHub search · MiniMax · commits · SilverKnightKMA/paseo-plugins · @SilverKnightKMA](https://github.com/SilverKnightKMA/paseo-plugins/commit/48b2440463e12941cc5eeae4ece206ecd2906ef7)
> now carry the FULL pi reference '<pi-provider>/<model>' (magpie/minimax/MiniMax-M3),
- 2026-11-13T19:05:09.193Z · [GitHub search · MiniMax · commits · SilverKnightKMA/paseo-plugins · @SilverKnightKMA](https://github.com/SilverKnightKMA/paseo-plugins/commit/72a2cedffcb6424e84392d7f17df3a9a5679bac0)
- 2026-11-13T19:05:09.193Z · [GitHub search · MiniMax · commits · rodrigueztaylor87/rod · @rodrigueztaylor87](https://github.com/rodrigueztaylor87/rod/commit/64189a80753eb37b84793d886ded01ab340508fb)

## Mga link sa ebidensiya

- [GitHub search · MiniMax · pulls · EvolvingLMMs-Lab/lmms-eval · @sahirema](https://github.com/EvolvingLMMs-Lab/lmms-eval/pull/1571)
- [GitHub search · MiniMax · commits · Lore-Hex/quill-router · @jperla](https://github.com/Lore-Hex/quill-router/commit/664e5541dd2f40274ca93378ac783395e1c3576c)
- [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 · pulls · MrGeva/vllm · @MrGeva](https://github.com/MrGeva/vllm/pull/3)
- [GitHub search · MiniMax · pulls · NVIDIA-NeMo/Megatron-Bridge · @pruprakash](https://github.com/NVIDIA-NeMo/Megatron-Bridge/pull/6839)
- [GitHub search · MiniMax · pulls · NVIDIA/TensorRT-LLM · @ziyixiong-nv](https://github.com/NVIDIA/TensorRT-LLM/pull/21438)
- [GitHub search · MiniMax · pulls · Pb314314/vllm · @Pb314314](https://github.com/Pb314314/vllm/pull/19)
- [GitHub search · MiniMax · pulls · Pinvou/pinvou-agent · @asto18089](https://github.com/Pinvou/pinvou-agent/pull/841)
- [GitHub search · MiniMax · pulls · Portkey-AI/models · @polease](https://github.com/Portkey-AI/models/pull/1032)
- [GitHub search · MiniMax · pulls · Quarkgluonmixture/quarkspace · @github-actions[bot]](https://github.com/Quarkgluonmixture/quarkspace/pull/130)
- [GitHub search · MiniMax · pulls · ROCm/ATOM · @rbrugaro-amd](https://github.com/ROCm/ATOM/pull/2701)
- [GitHub search · MiniMax · pulls · ROCm/ATOM · @yhl-amd](https://github.com/ROCm/ATOM/pull/3194)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @RElbers](https://github.com/ROCm/aiter/pull/7218)
- [GitHub search · MiniMax · pulls · ROCm/aiter · @zhiding512](https://github.com/ROCm/aiter/pull/7619)
- [GitHub search · MiniMax · commits · SilverKnightKMA/paseo-plugins · @SilverKnightKMA](https://github.com/SilverKnightKMA/paseo-plugins/commit/48b2440463e12941cc5eeae4ece206ecd2906ef7)
- [GitHub search · MiniMax · commits · SilverKnightKMA/paseo-plugins · @SilverKnightKMA](https://github.com/SilverKnightKMA/paseo-plugins/commit/72a2cedffcb6424e84392d7f17df3a9a5679bac0)
- [GitHub search · MiniMax · pulls · anomalyco/models.dev · @rjuez00](https://github.com/anomalyco/models.dev/pull/13443)
- [GitHub search · MiniMax · pulls · ashhart/TensorFold · @ashhart](https://github.com/ashhart/TensorFold/pull/3123)
- [GitHub search · MiniMax · pulls · boelabs/ai-gateway · @jhisdev](https://github.com/boelabs/ai-gateway/pull/22)
- [GitHub search · MiniMax · pulls · cbc02009/container-images · @mahoro-ai](https://github.com/cbc02009/container-images/pull/2)
- [GitHub search · MiniMax · pulls · ghuntley/how-to-build-a-coding-agent · @octo-patch](https://github.com/ghuntley/how-to-build-a-coding-agent/pull/20)
- [GitHub search · MiniMax · pulls · goonercbot/hermes-agent · @goonercbot](https://github.com/goonercbot/hermes-agent/pull/39)
- [GitHub search · MiniMax · pulls · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/pull/134)
- [GitHub search · MiniMax · pulls · inference-sim/blis-vllm · @sriumcp](https://github.com/inference-sim/blis-vllm/pull/99)
- [GitHub search · MiniMax · pulls · inference-sim/inference-sim · @sriumcp](https://github.com/inference-sim/inference-sim/pull/2104)
- [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6140)
- [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6141)
- [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6142)
- [GitHub search · MiniMax · pulls · iv-org/invidious · @Fijxu](https://github.com/iv-org/invidious/pull/6143)
- [GitHub search · MiniMax · pulls · jundot/omlx · @begcdn](https://github.com/jundot/omlx/pull/5239)
- [GitHub search · MiniMax · pulls · maximhq/bifrost · @octo-patch](https://github.com/maximhq/bifrost/pull/5184)
- [GitHub search · MiniMax · pulls · maximhq/bifrost · @jimaldon](https://github.com/maximhq/bifrost/pull/7124)
- [GitHub search · MiniMax · pulls · openJiuwen-ai/jiuwenswarm · @octo-patch](https://github.com/openJiuwen-ai/jiuwenswarm/pull/2117)
- [GitHub search · MiniMax · pulls · overnightworks/speech · @dependabot[bot]](https://github.com/overnightworks/speech/pull/354)
- [GitHub search · MiniMax · commits · rodrigueztaylor87/rod · @rodrigueztaylor87](https://github.com/rodrigueztaylor87/rod/commit/64189a80753eb37b84793d886ded01ab340508fb)
- [GitHub search · MiniMax · pulls · sgl-project/sglang · @alphabetc1](https://github.com/sgl-project/sglang/pull/50334)
- [GitHub search · MiniMax · pulls · tenstorrent/tt-metal · @mdragulaTT](https://github.com/tenstorrent/tt-metal/pull/64261)
- [GitHub search · MiniMax · pulls · tenstorrent/tt-metal · @mdragulaTT](https://github.com/tenstorrent/tt-metal/pull/64459)
- [GitHub search · MiniMax · pulls · thequanhcm/quanpt1 · @thequanhcm](https://github.com/thequanhcm/quanpt1/pull/28)
- [GitHub search · MiniMax · pulls · viztor/dsh-opencode-patch · @viztor](https://github.com/viztor/dsh-opencode-patch/pull/113)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @wolfcomos](https://github.com/vllm-project/vllm/pull/62098)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @afriedri](https://github.com/vllm-project/vllm/pull/64089)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @MrGeva](https://github.com/vllm-project/vllm/pull/65433)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @MrGeva](https://github.com/vllm-project/vllm/pull/65434)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @cjackal](https://github.com/vllm-project/vllm/pull/65453)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @Pb314314](https://github.com/vllm-project/vllm/pull/67338)
- [GitHub search · MiniMax · pulls · vllm-project/vllm · @Vincent-PRO-AI](https://github.com/vllm-project/vllm/pull/67518)
- [GitHub search · MiniMax · pulls · xiaoY233/Chat2API · @octo-patch](https://github.com/xiaoY233/Chat2API/pull/184)
- [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)

## Mga senyal