Unverified50% confidenceReleaseExact time
The developer uses the Qwen3.6-35B-A3B model: it has 35B weights in total, uses only 3B weights when processing each word, has a computational cost comparable to ordinary 4B and 8B models, and runs on their own gateway (vLLM).
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/13/2026
First Seen
Valid until: 1/11/2027
Sources
Related Entities
Related Claims
UnverifiedThe developer's demo used the Qwen 3.6 27B model, and users with limited hardware are advised to choose lighter 4B or 8B models.79% similarUnverifiedModels in the 3B–4B range, such as Qwen2.5 3B or Llama 3.2 3B, require only 2 to 2.5GB of VRAM under Q4 quantization.77% similarVerifiedQwen3-series coding models load at approximately 20 GB and can expand to 35–40 GB during inference.77% similarVerifiedThe Qwen2.5:7B model is approximately 4 GB in size after Q4 quantization.74% similarUnverifiedQwen2 0.5B模型（5亿参数）运行仅需约352MB显存73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1013083API
curl https://kongchang.com/api/v1/knowledge/claims/1013083MCP
get_claim(id=1013083)