Qwen3.8
Alibaba
時系列
- · 参照リンク · GitHub search · Alibaba · commits · onprem-ai/deployment-catalog · @tomas-polach
- · 参照リンク · GitHub search · Alibaba · commits · finereli/pyramid-mcp · @finereli
- · 参照リンク · GitHub search · Alibaba · commits · fdietze/pi-infinite-context · @fdietze
- · Alex の投稿 · シリーズ · GitHub search · Alibaba · commits · agentgateway/agentgateway · @github-actions[bot]
抜粋
json.providers.nebius.models["Qwen/Qwen3.8-27B"].name = "Qwen3.8 27B";
- · 参照リンク · GitHub search · Alibaba · commits · aabhay007/Personal-AI-Core · @aabhay007
- · 参照リンク · GitHub search · Alibaba · commits · QuantumKuba/Diffactly · @QuantumKuba
- · 参照リンク · GitHub search · Alibaba · commits · Hallo-ich-bin-PeppaPig/Singularity · @Hallo-ich-bin-PeppaPig
- · 参照リンク · GitHub search · Alibaba · commits · Azure/azureml-assets · @hanbitmyths
抜粋
Add MAI, Qwen3.8 and Nemotron 3.5 Foundry Local coding models (#5452)
- · Alex の投稿 · シリーズ · GitHub search · Alibaba · pulls · xadupre/devdash-agents · @xadupre
抜粋
- Real Qwen3.8 27B on H200-101, physical CUDA 7: streaming generation and a follow-up turn succeeded. Used a temporary configuration with 16 KV blocks and
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] qwen38_flash package entry: register() and READMEs
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] MTP draft model
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] KV offload pool regions
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Runner input adapter
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Hyper-connection layers
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] GDN layer input preparation
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Pooled KV cache materialization
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] PD transfer: identity, manifest and Raiden connector
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Config hook, geometry checks and pooled cache layout
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] PLE feature: tables, host lookup, storage, offload and pooled PLE
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Sequence-parallel prefill backbone
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] QSA attention layer
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Model-owned GDN op and layer (no register_oot)
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Per-group Mamba index accessors
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Mamba tail fold: export prefill checkpoints from inside the step
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] In-place rows update op
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] PLE Torch-to-JAX ops
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] QSA Torch-to-JAX ops
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Torch bridge for the QSA kernels
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] GDN input preparation kernel
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Unified KV pool kernels
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] PLE lookup and gate kernels
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Hyper-connection kernels
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Pooled QSA kernel over the unified KV pool
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] QSA KV write kernels (fused, main, compressed)
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] QSA union-walk kernels (plan, dense, SparseCore gather, prefill)
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] Qwen3.8-Flash model and registration hooks
- · 参照リンク · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
抜粋
[Model][Qwen3.8] QSA main-KV decode context parallelism over rank pairs (DCP-2) kernels
- · 参照リンク · GitHub search · Alibaba · pulls · vjmichael/Lu-AI-GUI · @vjmichael
抜粋
- Physical interactive owner Windows/Radeon Qwen3.8, Vulkan VRAM/TFFB TTFT p50/p95, and actual local installed MMProj; hosted CPU small models do not substitute.
- · 参照リンク · GitHub search · Alibaba · pulls · vitessio/vitess · @jdoupe
- · 参照リンク · GitHub search · Alibaba · pulls · nearai/cvm-compose-files · @PierreLeGuen
- · 参照リンク · GitHub search · Alibaba · pulls · michsindlinger/specwright · @michsindlinger
- · Alex の投稿 · シリーズ · GitHub search · Alibaba · pulls · SceneWorks/SceneWorks · @michaeltrefry
抜粋
SW-7119: Qwen3.8 27B describer model and tier in core, worker, API and catalog
- · Alex の投稿 · シリーズ · GitHub search · Alibaba · pulls · NVIDIA/hyperflare-inference · @Raydius
抜粋
The pinned Qwen3.8 Flash Next checkpoint trains a 262,144-token context (theta 10M, interleaved MRoPE 11/11/10). This adds a declared serving variant of the same weights
- · 参照リンク · GitHub search · Alibaba · pulls · FlashML-org/FreeToken · @zihaomu
- · Alex の投稿 · シリーズ · GitHub search · Alibaba · pulls · BuQQzz/SHIELD2.0 · @BuQQzz
抜粋
Makes Qwen3.8 27B (IQ1_M) a working coding model on an 8 GB card (RTX 2070 SUPER) through SHIELD's own llama-server.
- · 参照リンク · GitHub search · Alibaba · pulls · Azure/azureml-assets · @hanbitmyths
抜粋
Add MAI, Qwen3.8 and Nemotron 3.5 Foundry Local coding models
- · 参照リンク · GitHub search · Alibaba · pulls · 0xburden/modelship · @0xburden
抜粋
- Reviewed immutable catalog of 25 standalone Qwen3.8 GGUF quantizations; wizard displays sizes, numeric/letter/full-value selection and a visible UD-Q4_K_M default. No hardware-driven quant substitution; flags
- · 参照リンク · GitHub · QwenLM/qwen-code · PR
抜粋
- The Standard preset has the same qwen3.8 effort and GLM thinking gaps and will be aligned in a follow-up
- · Alex の投稿 · シリーズ · Alex Getman
抜粋
🚨 Qwen3.8 Omni Flash is out It can see, hear and understand video, then actually act on it Give it hours of footage and it can find the right moments, translate, edit or turn it into notes. Give it a meeting and it can pull out tasks and start working on them 1M context, API from $0.43
- · Alex の投稿 · シリーズ · Alex Getman
抜粋
🚨 QWEN3.8 FLASH NEXT IS OUT The leaked specs were real: • 125B multimodal MoE • Only 6B active • 262K native context, up to 1M Qwen says it beats Claude Opus 4.6 across coding, agentic tasks, tool use and several multimodal benchmarks.
- · Alex の投稿 · シリーズ · Alex Getman
抜粋
🚨 QWEN3.8 FLASH NEXT DROPS TODAY A ModelScope description that was quickly taken down reportedly revealed: • 125B main parameters • 51B N gram embeddings • Only 6B active per token • Multimodal MoE architecture The model is also an official preview of the upcoming Qwen4 architecture.
- · Alex の投稿 · Alex Getman
抜粋
⚡️ Qwen 3.8 27B beat GPT 5.6 Terra This happened in the AA Agentic Index, a benchmark that tests how well models handle agentic tasks. And you can run Qwen 3.8 27B locally on a single RTX 3090.
- · Alex の投稿 · Alex Getman
抜粋
⚡️ Alibaba released Qwen3.8’s weights A model at Claude Opus 4.6 level, a flagship just six months ago, can now run on a single RTX 4090. Qwen3.8 27B uses around 17 GB of VRAM when quantized and should deliver ~50 tok/s.
- · Alex の投稿 · Alex Getman
抜粋
Alibaba just teased Qwen3.8 with 2.4T parameters This is the new Qwen flagship, with open weights coming soon. Alibaba ranks it right below Claude Fable 5. Public benchmarks are still missing, so nobody really knows how close it actually is to Fable, or how it stacks up against GPT-5.6 Sol and Kimi K3. Pricing is the fun part. Kimi K3 costs $3/$15, so Qwen3.8 will probably land in the same range. The preview is already live in Qoder. It uses 10x fewer credits during the day and 50x fewer at night, so you can basically test it for free.
仕様
レーダー · Qwen3.8 · リリース日未確認
よくある質問
Qwen3.8 のリリース日は？
リリース日未確認