[KongchangAI]
Unverified50% confidenceBenchmarkExact time

微软公开的基准测试显示BitNet b1.58在困惑度等指标上已接近同规模全精度模型,但这一结论在独立研究者中仍存争议

1
Sources
50%
Confidence
Long-term
Relevance
7/17/2026
First Seen

Sources

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/544061
API
curl https://kongchang.com/api/v1/knowledge/claims/544061
MCP
get_claim(id=544061)