LongCat 2.5 Release - MathVision - w/ tools: leaderboard
LongCat-2.5 final release detailed native result table, November 2026. External:true cells are explicitly cited other-evaluator results and excluded from this component; self-run numeric cells retained. No Preview/final or release-generation aliases. Detailed table used instead of inconsistent rounded headline charts. CharXiv-RQ (w/ tools) and MathVision (w/ tools): Temperature = 1.0 and top_p = 0.95, with a maximum generation length of 131,972 tokens. Metric: Score (%).
Metric: Score (%). Source: longcat.ai. Saturation forecast: Around February 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4.1 Flash | 97.4 | #37 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=longcat-2-5-release-mathvision-w-tools · How It Works · Data refreshed daily, snapshot 2026-11-05.