[KongchangAI]
Unverified50% confidenceBenchmarkExact time

The eight evaluated models generally have lower accuracy in Bengali than in English, and among them Mistral-Nemo has the smallest cross-lingual accuracy gap, at 5.12 percentage points.

1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen

Sources

Related Entities

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/1154785
API
curl https://kongchang.com/api/v1/knowledge/claims/1154785
MCP
get_claim(id=1154785)