List your company

Early-stopping budget estimation on Search-R1, Sokoban, SWE-bench, and Warehouse Aggregate (test)

2.2False-Abort RateClaude Opus 4.7
2.0243.2124.45.588May 29, 2026

Evaluation results

MethodLinks
Claude Opus 4.7Model=Claude Opus 4.72026.052.228.25062
Gemini 3.1 ProModel=Gemini 3.1 Pro2026.052.855.763123
Claude Sonnet 4.6Model=Claude Sonnet 4.62026.053.349.676101
Qwen3 235BModel=Qwen3 235B2026.054.938.8190140
GPT-5.2 InstantModel=GPT-5.2 Instant2026.056.664.1183124