List your company

Web Agent Task Completion on HLE

37.6AccuracySlimSearcher (SFT+RL)
-0.5689.34119.2529.159Jun 5, 2026

Evaluation results

MethodLinks
SlimSearcher (SFT+RL)Backbone Policy=Tongyi-DeepResearch, Optimization Strategy=SFT+RL2026.0637.619.8211,418
SlimSearcher (SFT)Backbone Policy=Tongyi-DeepResearch, Optimization Strategy=SFT2026.0636.627.1612,936
Tongyi-DeepResearch2026.0635.823.9213,664
Prompt ControlBackbone Policy=Tongyi-DeepResearch, Optimization Strategy=Prompt Control2026.0634.923.9114,107
WebLeaperType=Open-weights Agent2026.0629.824.8611,924
SlimSearcher (SFT+RL)Backbone Policy=Qwen3-30B-A3B-Instruct-2507, Optimization Strategy=SFT+RL2026.0627.819.5111,497
OpenAI DeepResearchType=Proprietary System2026.0626.6——
SlimSearcher (SFT)Backbone Policy=Qwen3-30B-A3B-Instruct-2507, Optimization Strategy=SFT2026.0625.927.8613,169
OpenAI o3Type=Proprietary System2026.0624.9——
Claude-4-SonnetType=Proprietary System2026.0620.3——
Qwen3-235B-A22B-InstructType=Open-weights Agent2026.0619.97.452,960
Kimi-K2-Instruct-0905Type=Open-weights Agent2026.0614.65.172,349
WebExplorerType=Open-weights Agent2026.0611.615.526,579
DeepSeek-V3Type=Open-weights Agent2026.066132,713
Qwen3-30B-A3B-Instruct-2507Type=Open-weights Agent, Unified Environment=true2026.060.925.772,570