Web Browsing on BrowseComp
85.9AccuracyGemini-3.1-Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3.1-ProCategory=Proprietary Agents2026.05 | 85.9 | |
| Claude-4.6-OpusCategory=Proprietary Agents2026.05 | 83.7 | |
| DeepSeek-V4-Pro-MaxCategory=Open-Source Agents2026.05 | 83.4 | |
| Kimi-K2.6Category=Open-Source Agents2026.05 | 83.2 | |
| Qwen3.5-397B-A17BCategory=Open-Source Agents2026.05 | 78.6 | |
| Seed-2.0-ProCategory=Proprietary Agents2026.05 | 77.3 | |
| GLM-5.0Category=Open-Source Agents2026.05 | 75.9 | |
| Argus-35B-A3B (Parallel)Category=Parallel Agents, Backbone=35B-A3B, Mode=Parallel2026.05 | 74.5 | |
| MiroThinker-1.7Category=Open-Source Deep Research Agents2026.05 | 74 | |
| PassK=82026.04 | 73.33 | |
| AggAgentK=82026.04 | 71.33 | |
| BoNK=82026.04 | 70.67 | |
| SolAggK=82026.04 | 70.67 | |
| SummAggK=82026.04 | 70 | |
| Tongyi-DeepResearch HeavyCategory=Parallel Agents2026.05 | 69 | |
| PassK=42026.04 | 68.64 | |
| PassK=8, Model=Qwen3.5-122B, Aggregation Strategy=None2026.04 | 67.33 | |
| AggAgentK=42026.04 | 67.33 | |
| FewToolK=82026.04 | 67.33 | |
| SummAggK=42026.04 | 66.67 | |
| WMVK=82026.04 | 66.67 | |
| BoNK=42026.04 | 66.06 | |
| GPT-5.2Category=Proprietary Agents2026.05 | 65.8 | |
| WMVK=42026.04 | 64.73 | |
| SolAggK=42026.04 | 64.44 | |
| AggAgentK=8, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 62.67 | |
| FewToolK=42026.04 | 62.63 | |
| Argus-35B-A3B (Solo)Category=Parallel Agents, Backbone=35B-A3B, Mode=Solo2026.05 | 62.2 | |
| SummAggK=8, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 62 | |
| SolAggK=8, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 61.33 | |
| PassK=22026.04 | 60.86 | |
| MVK=82026.04 | 60.67 | |
| AggAgentK=22026.04 | 60.22 | |
| WMVK=8, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 60 | |
| SolAggK=22026.04 | 60 | |
| SummAggK=22026.04 | 59.78 | |
| PassK=4, Model=Qwen3.5-122B, Aggregation Strategy=None2026.04 | 59.53 | |
| WMVK=22026.04 | 59.4 | |
| BoNK=22026.04 | 59.36 | |
| BoNK=8, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 58.67 | |
| FewToolK=22026.04 | 57.38 | |
| MVK=42026.04 | 56.58 | |
| AggAgentK=4, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 56.22 | |
| SummAggK=4, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 56 | |
| FewToolK=8, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 56 | |
| SolAggK=4, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 55.78 | |
| Searcher-35B-A3BCategory=Parallel Agents, Backbone=35B-A3B2026.05 | 55 | |
| BoNK=4, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 54.83 | |
| WMVK=4, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 54.3 | |
| GLM-4.5 HeavyCategory=Parallel Agents2026.05 | 54 | |
| FewToolK=4, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 51.42 | |
| MVK=22026.04 | 50.4 | |
| PassK=2, Model=Qwen3.5-122B, Aggregation Strategy=None2026.04 | 50.38 | |
| PassK=12026.04 | 50.17 | |
| SummAggK=2, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 49.33 | |
| AggAgentK=2, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 48.89 | |
| SolAggK=2, Model=Qwen3.5-122B, Aggregation Strategy=LLM-based Aggregation2026.04 | 48.44 | |
| BoNK=2, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 48.05 | |
| WMVK=2, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 47.79 | |
| MVK=8, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 47.33 | |
| FewToolK=2, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 46.21 | |
| MVK=4, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 45.54 | |
| Tongyi-DeepResearchCategory=Open-Source Deep Research Agents2026.05 | 43.4 | |
| Qwen3.5-35B-A3BCategory=Open-Source Agents2026.05 | 42.1 | |
| MVK=2, Model=Qwen3.5-122B, Aggregation Strategy=Heuristic Aggregation2026.04 | 40.74 | |
| PassK=1, Model=Qwen3.5-122B, Aggregation Strategy=None2026.04 | 40.25 | |
| PivotRLBase Model=Nemotron-3-Super, Post-training Stage=After RL Stage2026.03 | 25.04 | |
| Nemotron-3-SuperPost-training Stage=SFT2026.03 | 13.03 |