Unverified50% confidenceFactTime unknown
Qwen 3.6 35B A3B完全开源，权重在Hugging Face上可用，大小约为72GB
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
Verified阿里巴巴推出Qwen 3.6 35B A3B，拥有350亿参数的混合专家模型，同一时间只有30亿参数处于激活状态，权重约72GB已在Hugging Face开源78% similarUnverifiedQwen3系列模型可通过Hugging Face获取，支持128K或32K上下文长度77% similarUnverifiedQwen-7B经过4-bit量化后，一张搭载24GB显存的消费级显卡即可流畅运行74% similarUnverified运行具备实用能力的开源模型（如Llama 3或Qwen2.5）通常需要至少16GB显存的GPU71% similarUnverified3B～4B级别模型如Qwen2.5 3B或Llama 3.2 3B在Q4量化下仅需2～2.5GB显存71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/55789API
curl https://kongchang.com/api/v1/knowledge/claims/55789MCP
get_claim(id=55789)