BrowseComp — Explicit single-agent scaffold, distinct tools axis from multi-agent/Agent-Swarm siblings
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.8 | Anthropic | 88.5 | 2 | 2026-06-12 |
| 2 | Claude Mythos 5 | Anthropic | 88 | 2 | 2026-06-12 |
| 3 | GPT-5.5 | OpenAI | 84.4 | 1 | 2026-08-12 |
| 4 | Claude Opus 4.6 | Anthropic | 83.7 | 1 | 2026-06-03 |
| 5 | Claudes | Anthropic | 76.2 | 1 | 2026-08-12 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.