UTC

Qwen3.8 2.4T A95B

Alibaba (Qwen) · open weights · served by 29 providers · tool calling, reasoning

alibaba/qwen3.8-2.4t-a95b

Weights
Open
Released
2026-08-12
Knowledge cutoff
Not stated
Context window
1M
Deprecation dates

When each platform deprecates Qwen3.8 2.4T A95B, as LiteLLM records it. Plan the move before the date: after it, the platform may stop serving the model or stop updating it.

  • Together AIdeprecated 2026-11-04together_ai/Qwen/Qwen3.8-2.4T-A95Bsuccessor: together_ai/Qwen/Qwen3.7-Max
Every upcoming deprecation →
Hugging Face repository

Qwen/Qwen3.8-2.4T-A95B at commit 4f885e51d6, last changed 2026-11-07

Licence
qwen3.8-max
Access
Open download
Library
transformers
Formats
.safetensors
  • No model signature file is published with the weights.

What the Hub's scanners report

43 weight files, 0 with a scan result. No scanner reports an issue with them.

The model at this commit, its 43 weight files with their SHA-256.

Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.

Who serves it
  • AbliteratedLLMqwen3.8-2.4t-a95b
  • AIHubMixqwen3.8-2.4t-a95b
  • AI/ML APIalibaba/qwen3.8-2.4t-a95b
  • Cortecsqwen3.8-2.4t-a95b
  • Deep InfraQwen/Qwen3.8-2.4T-A95Bdeprecated
  • Eden AIqwen/qwen3.8-2.4t-a95b
  • Fireworks AIaccounts/fireworks/models/qwen3p8-2p4t-a95b
  • Hugging FaceQwen/Qwen3.8-2.4T-A95B
  • Charm Hyperqwen3.8-2.4t-a95b
  • IteraComputeqwen/qwen3.8-2.4t-a95b
  • Kelnqwen/qwen3.8-2.4t-a95b
  • Kilo Gatewayqwen/qwen3.8-2.4t-a95b
  • DevPass (LLM Gateway)qwen3.8-2.4t-a95b
  • LLM Gatewayalibaba/qwen3.8-2.4t-a95b
  • LLM Gatewayfireworks/qwen3.8-2.4t-a95b
  • LLM Gatewaynovita/qwen3.8-2.4t-a95b
  • Lyceumqwen/qwen3.8-2.4t-a95b
  • Merge Gatewayqwen/qwen3.8-2.4t-a95b
  • ModalQwen/Qwen3.8-2.4T-A95B
  • Novita AIqwen/qwen3.8-2.4t-a95b
  • OpenRouterqwen/qwen3.8-2.4t-a95b
  • Opperqwen3.8-2.4t-a95b
  • Requestyqwen3.8-2.4T-A95B
  • Requestyqwen3.8-2.4T-A95B@eu
5 more providers

RunInfra (Inferact/Qwen3.8-2.4T-A95B-NVFP4), SiliconFlow (Qwen/Qwen3.8-2.4T-A95B), TensorX (qwen/qwen3.8-2.4t-a95b), Venice AI (qwen-3-8-2-4t-a95b), Vercel AI Gateway (alibaba/qwen3.8-2.4t-a95b)

Running it yourself

Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.

How supply-chain attacks on models work →

Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.