# BrowseComp — AI model rankings

Can the AI browse the web and track down hard-to-find answers? Higher is better.

17 tracked models have a published BrowseComp score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Kimi K3 | Moonshot AI | 91.2% | Jul 16 2026 |
| 2 | Claude Opus 5 | Anthropic | 90.8% | Jul 24 2026 |
| 3 | GPT-5.5-Pro | OpenAI | 90.1% | Apr 23 2026 |
| 4 | GPT-5.4-Pro | OpenAI | 89.3% | Mar 5 2026 |
| 5 | Claude Fable 5 | Anthropic | 86.9% | Jun 9 2026 |
| 6 | Gemini 3.1 Pro | Google | 85.9% | Feb 19 2026 |
| 7 | Claude Sonnet 5 | Anthropic | 84.7% | Jun 30 2026 |
| 8 | GPT-5.5 | OpenAI | 84.4% | Apr 23 2026 |
| 9 | Claude Opus 4.8 | Anthropic | 84.3% | May 28 2026 |
| 10 | Claude Opus 4.6 | Anthropic | 83.7% | Feb 5 2026 |
| 11 | DeepSeek-V4-Pro | DeepSeek | 83.4% | Apr 24 2026 |
| 12 | GPT-5.4 | OpenAI | 82.7% | Mar 5 2026 |
| 13 | Claude Opus 4.7 | Anthropic | 79.3% | Apr 16 2026 |
| 14 | GLM-5 | Z.ai | 75.9% | Feb 12 2026 |
| 15 | Qwen3.5 | Qwen | 69% | Feb 16 2026 |
| 16 | GLM-5.1 | Z.ai | 68% | Apr 7 2026 |
| 17 | GLM-4.7 | Z.ai | 52% | Dec 22 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/browsecomp
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Benchmark scores are the figures published by the releasing lab at launch.