Unverified50% confidenceBenchmarkExact time
Mistral officially says that after three additional weeks of training, Large 4's Terminal Bench score rose from 28% to 41%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/30/2026
First Seen
Valid until: 1/28/2027
Sources
Related Entities
Related Claims
UnverifiedOn TerminalBench 4.0, Large 4 scores 41%, a 13-percentage-point improvement over the preview version from three weeks earlier, and Mistral says it ranks first among open-source models.70% similarUnverifiedAfter fine-tuning on the MindView training set, BAGEL's overall score improved by 24.3 points.64% similarUnverifiedThe Terminal Bench 4.0 score rose from 31.2 to 42.3, an improvement of 11.1 points.63% similarUnverifiedThe Program Bench score rose from 20.3 to 29.1, an improvement of 8.8 points, using the 'Almost' criterion (passing 5 of 9 tests counts as a pass).63% similarUnverifiedMistral Large 4's AA score is 38, a significant improvement over the previous generation Mistral Medium 3.5's score of 14.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114964API
curl https://kongchang.com/api/v1/knowledge/claims/1114964MCP
get_claim(id=1114964)