Unverified60% confidenceBenchmarkExact time
Mistral Large 4 uses an average of 47% fewer tokens per task on equivalent tasks.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
10/31/2026
First Seen
Valid until: 1/29/2027
Sources
Mistral Opens Trillion-Parameter Model, OpenAI Agent Monitoring, and Claude Code Controversy
bilibili不评论绝不在p站评论10/30/2026
Related Entities
Related Claims
VerifiedMistral Large 4 narrowly outperformed GPT-6 Astra on visual grounding tasks with a score of 42% versus 41%.74% similarUnverifiedThe new version of Mistral Large 4 used 30% fewer tokens in its output.73% similarUnverifiedOn the "can the change be merged" test, Mistral Large 4 scored 19.7%, only 0.4 percentage points above second place.70% similarUnverifiedMistral Large 4 scored approximately 62% on the real-world software engineering benchmark DeepSwee.69% similarUnverifiedMistral Large 4's score on the terminal coding test rose from 28% in the preview version to 41%, an increase of 13 percentage points.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1119744API
curl https://kongchang.com/api/v1/knowledge/claims/1119744MCP
get_claim(id=1119744)