Unverified50% confidenceFactExact time
A smaller 4B model can serve as an excellent skill-writing 'teacher' for a 27B large model.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn the experiment, the 27B model scored 61.0% on LiveMath when skills were written by the 4B model, whereas it scored 56.3% when it wrote the skills itself65% similarUnverified评测者认为通义千问3.6 27B是30B级别中对高考数学题表现最好的模型62% similarUnverifiedThe vocabulary size obtained by the model after training the tokenizer was 3,266.58% similarUnverified4B表示模型参数量为40亿，属于小型语言模型（SLM）范畴57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/829755API
curl https://kongchang.com/api/v1/knowledge/claims/829755MCP
get_claim(id=829755)