UTC

Qwen3.6 35B-A3B

Alibaba (Qwen) · open weights · served by 50 providers · tool calling, reasoning

alibaba/qwen3.6-35b-a3b

Weights
Open
Released
2026-04-17
Knowledge cutoff
2025-04
Context window
262K
Hugging Face repository

Qwen/Qwen3.6-35B-A3B at commit 345d467b11, last changed 2026-11-07

Licence
apache-2.0
Access
Open download
Library
transformers
Formats
.safetensors
  • No model signature file is published with the weights.

What the Hub's scanners report

26 weight files, 26 with a scan result. No scanner reports an issue with them.

The model at this commit, its 26 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
  • 302.AIqwen3.6-35b-a3b
  • AIHubMixqwen3.6-35b-a3b
  • Ambientqwen/qwen3.6-35b-a3b
  • Boundlessqwen3.6
  • Cortecsqwen3.6-35b-a3b
  • Deep InfraQwen/Qwen3.6-35B-A3B
  • DoublewordQwen/Qwen3.6-35B-A3B-FP8
  • EmpirioLabs AIqwen3-6-35b-a3b
  • engyqwen3.6-35b-a3b
  • evrocQwen/Qwen3.6-35B-A3B
  • GreenPTqwen3.6-35b-a3b
  • HetznerQwen/Qwen3.6-35B-A3B-FP8beta
  • Hugging FaceQwen/Qwen3.6-35B-A3B
  • InferXQwen3.6-35B-A3B-FP8
  • InferXQwen3.6-35B-A3B-fp8-no-thinking
  • Kilo Gatewayqwen/qwen3.6-35b-a3b
  • Klavo AIqwen/qwen3.6-35b-a3b
  • DevPass (LLM Gateway)qwen3.6-35b-a3b
  • LLM Gatewayalibaba/qwen3.6-35b-a3b
  • LLM Gatewaynovita/qwen3.6-35b-a3b
  • LLMTRqwen3-6-35b
  • Mallard AIqwen/qwen3.6-35b-a3b
  • Merge Gatewayqwen/qwen3.6-35b-a3b
  • NaNqwen3.6
26 more providers

NanoGPT (TEE/qwen3.6-35b-a3b), NanoGPT (qwen/qwen3.6-35b-a3b), NanoGPT (qwen/qwen3.6-35b-a3b:thinking), NEAR AI Cloud (Qwen/Qwen3.6-35B-A3B-FP8), Neuralwatt (qwen3.6-35b), Neuralwatt (qwen3.6-35b-fast), Neuralwatt (qwen3.6-35b-flex), Novita AI (qwen/qwen3.6-35b-a3b), OpenRouter (qwen/qwen3.6-35b-a3b), Opper (qwen3.6-35b-a3b), OrcaRouter (qwen/qwen3.6-35b-a3b), Phala (qwen/qwen3.6-35b-a3b), QVAC (qwen3.6-35b-a3b), SaladCloud AI Gateway (qwen3.6-35b-a3b), Scaleway (qwen3.6-35b-a3b), SiliconFlow (Qwen/Qwen3.6-35B-A3B), SiliconFlow (China) (Qwen/Qwen3.6-35B-A3B), SOKKAN Router (qwen/qwen3.6-35b-a3b-eu), Thinking Machines (Qwen/Qwen3.6-35B-A3B), Thinking Machines (Qwen/Qwen3.6-35B-A3B:peft:262144), TrustedTokens (Qwen/Qwen3.6-35B-A3B-FP8), Umans AI Coding Plan (umans-qwen3.6-35b-a3b), Venice AI (qwen3-6-35b-a3b), Vercel AI Gateway (alibaba/qwen3.6-35b-a3b), CoreWeave (Qwen/Qwen3.6-35B-A3B), ZeroGPU (qwen3.6-35b-a3b)

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.