Unverified50% confidenceBenchmarkExact time
Running Terminal Bench on the same model under different harnesses produced scores ranging from 69.8% to 74.2%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/17/2026
First Seen
Valid until: 12/16/2026
Sources
Related Entities
Related Claims
UnverifiedOn the Terminal Bench 2.1 benchmark, Ultra mode approaches 92%, surpassing competitors' performance of 88%.73% similarUnverifiedAt the mention level, the model achieved an overall precision of 74.1% and a recall of 70.5%.67% similarUnverifiedA benchmark score of 73.4% means that approximately one quarter of tasks still fall short, and this score was achieved under maximum compute.63% similarUnverifiedAt this compression ratio, the model retains 74.2% top-1% accuracy62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/928964API
curl https://kongchang.com/api/v1/knowledge/claims/928964MCP
get_claim(id=928964)