AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
Model reports
BrowseComp
Live web. The result depends on what the internet contained on the day of the run.
Evaluation context and evidence
- Source
- Model reports
- Publisher
- Not recorded
- Benchmark release
- 2025-04-10
- Collected
- Not recorded
- Modality
- Not recorded
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 9
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| DeepSeek-V4-Pro | 80.4 | percent | think high, pass@1 | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Pro | 83.4 | percent | think max, pass@1 | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Flash | 73.2 | percent | think max, pass@1 | 2026-06-25 · document_publication | DeepSeek-V4 |
| Gemini 3.1 Pro | 85.9 | percent | Thinking (High), Search + Python + Browse | 2026-02-19 · document_publication | Gemini 3.1 Pro |
| Kimi K3 | 91.2 | percent | reasoning=max | 2026-06-13 · document_publication | Kimi K3 |
| Qwen3.5-397B-A17B | 78.6 | percent | thinking (temp 0.6, top_p 0.95, top_k 20), discard-all context strategy | 2026-02-16 · document_publication | Qwen3.5-397B-A17B |
| Qwen3.5-397B-A17B | 69.0 | percent | thinking (temp 0.6, top_p 0.95, top_k 20), simple context-folding | 2026-02-16 · document_publication | Qwen3.5-397B-A17B |
| GLM-5.1 | 68.0 | percent | no context management, as reported | 2026-04-03 · document_publication | GLM-5.1 |
| GLM-5 | 62.0 | percent | no context management, retain most recent 5 turns | 2026-02-11 · document_publication | GLM-5 |