Qwen3.8
Alibaba
Timeline
- · Evidence links · GitHub search · Alibaba · commits · onprem-ai/deployment-catalog · @tomas-polach
- · Evidence links · GitHub search · Alibaba · commits · finereli/pyramid-mcp · @finereli
- · Evidence links · GitHub search · Alibaba · commits · fdietze/pi-infinite-context · @fdietze
- · Alex's posts · model family · GitHub search · Alibaba · commits · agentgateway/agentgateway · @github-actions[bot]
Excerpt
json.providers.nebius.models["Qwen/Qwen3.8-27B"].name = "Qwen3.8 27B";
- · Evidence links · GitHub search · Alibaba · commits · aabhay007/Personal-AI-Core · @aabhay007
- · Evidence links · GitHub search · Alibaba · commits · QuantumKuba/Diffactly · @QuantumKuba
- · Evidence links · GitHub search · Alibaba · commits · Hallo-ich-bin-PeppaPig/Singularity · @Hallo-ich-bin-PeppaPig
- · Evidence links · GitHub search · Alibaba · commits · Azure/azureml-assets · @hanbitmyths
Excerpt
Add MAI, Qwen3.8 and Nemotron 3.5 Foundry Local coding models (#5452)
- · Alex's posts · model family · GitHub search · Alibaba · pulls · xadupre/devdash-agents · @xadupre
Excerpt
- 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
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] qwen38_flash package entry: register() and READMEs
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] MTP draft model
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] KV offload pool regions
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Runner input adapter
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Hyper-connection layers
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] GDN layer input preparation
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Pooled KV cache materialization
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] PD transfer: identity, manifest and Raiden connector
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Config hook, geometry checks and pooled cache layout
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] PLE feature: tables, host lookup, storage, offload and pooled PLE
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Sequence-parallel prefill backbone
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] QSA attention layer
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Model-owned GDN op and layer (no register_oot)
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Per-group Mamba index accessors
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Mamba tail fold: export prefill checkpoints from inside the step
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] In-place rows update op
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] PLE Torch-to-JAX ops
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] QSA Torch-to-JAX ops
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Torch bridge for the QSA kernels
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] GDN input preparation kernel
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Unified KV pool kernels
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] PLE lookup and gate kernels
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Hyper-connection kernels
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Pooled QSA kernel over the unified KV pool
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] QSA KV write kernels (fused, main, compressed)
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] QSA union-walk kernels (plan, dense, SparseCore gather, prefill)
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] Qwen3.8-Flash model and registration hooks
- · Evidence links · GitHub search · Alibaba · pulls · vllm-project/vllm-torchtpu · @DRZDONG
Excerpt
[Model][Qwen3.8] QSA main-KV decode context parallelism over rank pairs (DCP-2) kernels
- · Evidence links · GitHub search · Alibaba · pulls · vjmichael/Lu-AI-GUI · @vjmichael
Excerpt
- 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.
- · Evidence links · GitHub search · Alibaba · pulls · vitessio/vitess · @jdoupe
- · Evidence links · GitHub search · Alibaba · pulls · nearai/cvm-compose-files · @PierreLeGuen
- · Evidence links · GitHub search · Alibaba · pulls · michsindlinger/specwright · @michsindlinger
- · Alex's posts · model family · GitHub search · Alibaba · pulls · SceneWorks/SceneWorks · @michaeltrefry
Excerpt
SW-7119: Qwen3.8 27B describer model and tier in core, worker, API and catalog
- · Alex's posts · model family · GitHub search · Alibaba · pulls · NVIDIA/hyperflare-inference · @Raydius
Excerpt
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
- · Evidence links · GitHub search · Alibaba · pulls · FlashML-org/FreeToken · @zihaomu
- · Alex's posts · model family · GitHub search · Alibaba · pulls · BuQQzz/SHIELD2.0 · @BuQQzz
Excerpt
Makes Qwen3.8 27B (IQ1_M) a working coding model on an 8 GB card (RTX 2070 SUPER) through SHIELD's own llama-server.
- · Evidence links · GitHub search · Alibaba · pulls · Azure/azureml-assets · @hanbitmyths
Excerpt
Add MAI, Qwen3.8 and Nemotron 3.5 Foundry Local coding models
- · Evidence links · GitHub search · Alibaba · pulls · 0xburden/modelship · @0xburden
Excerpt
- 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
- · Evidence links · GitHub · QwenLM/qwen-code · PR
Excerpt
- The Standard preset has the same qwen3.8 effort and GLM thinking gaps and will be aligned in a follow-up
- · Alex's posts · model family · Alex Getman
Excerpt
🚨 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's posts · model family · Alex Getman
Excerpt
🚨 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's posts · model family · Alex Getman
Excerpt
🚨 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's posts · Alex Getman
Excerpt
⚡️ 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's posts · Alex Getman
Excerpt
⚡️ 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's posts · Alex Getman
Excerpt
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.
Alex's take
Specifications
Radar · Qwen3.8 · Release date unknown
Questions
When was Qwen3.8 released?
Release date unknown