BrowseComp Long Context 128k: leaderboard
Long-context BrowseComp variant for hard-to-find web questions, evaluating whether models can use a 128k context budget to retrieve and reason over evidence.
Metric: Score (%). Source: openai.com. Source: llm-stats.com. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 92 | #110 |
| 2 | GPT-5 | 90 | #98 |
| 3 | GPT-5.1 | 90 | #135 |
| 4 | GPT-5.1 Instant | 90 | #284 |
| 5 | GPT-5.1 (Thinking) | 90 | #135 (GPT-5.1) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=browsecomp-long-context-128k · How It Works · Data refreshed daily, snapshot 2026-10-20.