Unverified50% confidenceFactExact time
The Jeff project distilled and fine-tuned two base models, Qwen3.5 and Gemma 4, into decision models with 0.8B and 2B parameters respectively.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/30/2026
First Seen
Valid until: 12/29/2026
Sources
Related Entities
Related Claims
UnverifiedQwen, LLaMA, and ChatGLM are base models that can be used for fine-tuning.74% similarUnverifiedThe author used the Qwen3-1.7B model for both generation and attribution analysis tasks.70% similarUnverifiedThe Qwen model can be deployed via LM Studio and used for local inference.70% similarVerified混合专家架构MoE通过稀疏激活机制，推理时只激活模型总参数的1/8至1/467% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/962961API
curl https://kongchang.com/api/v1/knowledge/claims/962961MCP
get_claim(id=962961)