Agentic
BrowseComp
Web browsing comprehensive benchmark.
6 modelos publicaron score
| # | Modelo | Empresa | Score |
|---|---|---|---|
| 1 | GPT-5.5 Pro | OpenAI | 90.1 |
| 2 | Claude Mythos Preview | Anthropic | 86.9 |
| 3 | Gemini 3.1 Pro | Google DeepMind | 85.9 |
| 4 | GPT-5.5 | OpenAI | 84.4 |
| 5 | Kimi K2.6 | Moonshot AI | 83.2 |
| 6 | Claude Opus 4.7 | Anthropic | 79.3 |