Unverified50% confidenceBenchmarkExact time
Mistral Small 4 scored only 0.699 on unit conversion inputs and produced 265 errors that differed from the correct label by exactly a power of ten.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/23/2026
First Seen
Valid until: 12/22/2026
Sources
Related Entities
Related Claims
UnverifiedThe question setter placed 256 (the result of 4^4 without adding 3) as a distractor option, differing from 259 by only 3.72% similarUnverifiedRegardless of the encoding scheme used, 4-bit precision can only distinguish 16 different numerical levels.67% similarUnverifiedINT4量化将原本由65536个不同值（FP16）表示的权重压缩到仅16个离散级别66% similarUnverifiedOn the held-out test set, FakeSpotter achieved a macro F1 score of 0.788 for short texts and 0.793 for long texts.63% similarUnverifiedEarly testing on the Opus model showed an exact string recognition rate of 0 out of 15, whereas after the release of Fable 5 the same test yielded 13 out of 15.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/942851API
curl https://kongchang.com/api/v1/knowledge/claims/942851MCP
get_claim(id=942851)