Chart reasoning
CharXiv Reasoningwith tools
The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better.
Rankings
Higher is betterCharXiv Reasoning — frequently asked questions
- What is CharXiv Reasoning?
- The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better.
- Which AI model scores highest on CharXiv Reasoning?
- GLM-5.3-Flash by Z.ai holds the best CharXiv Reasoning (with tools) result among tracked models, at 89.4% (released Aug 26 2026). Higher scores are better on this benchmark.
- What are the top 1 models on CharXiv Reasoning?
- 1. GLM-5.3-Flash (Z.ai) — 89.4%.
- How many models have a published CharXiv Reasoning score?
- 1 tracked model has a published CharXiv Reasoning (with tools) score. Scores are the figures reported by each lab at that model's release, so this page is a record of results over time rather than a re-run leaderboard.
- What is the best open model on CharXiv Reasoning?
- GLM-5.3-Flash by Z.ai is the highest-ranked model with downloadable weights on CharXiv Reasoning (with tools), scoring 89.4% at rank 1 overall.