Unverified50% confidenceBenchmarkExact time
After fine-tuning on agent trajectories, CLM achieved 87.6% on Terminal-Bench 2.1 and 81.6% on DeepSWE.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/24/2026
First Seen
Valid until: 12/23/2026
Sources
Related Entities
Related Claims
UnverifiedZero-shot Jev scores approximately 71% on the DeepSWE benchmark, which is lower than the fine-tuned CLM's 81.6%.65% similarUnverifiedUnder the DeepSuite V1.1 benchmark, the same model's scores fluctuate from 69.8% to 74.2% across different harnesses.65% similarUnverifiedWith context management enabled, Iris-mini achieves 82.2%, 84.8%, 86.9%, and 52.3% on BrowseComp, BrowseComp-ZH, DeepSearchQA, and HLE respectively.63% similarUnverifiedRunning Terminal Bench on the same model under different harnesses produced scores ranging from 69.8% to 74.2%.60% similarUnverifiedWhen SWE-bench was first released, the pass rate of top models was below 2%; the pass rate of leading systems has now surpassed 50%.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/946807API
curl https://kongchang.com/api/v1/knowledge/claims/946807MCP
get_claim(id=946807)