A scientific chart reasoning benchmark that tests whether models can understand, interpret, and reason about complex scientific visualizations including plots, diagrams, and data charts.
BenchLM mirrors the published score view for CharXiv. Claude Mythos Preview leads the public snapshot at 93.2% , followed by Muse Spark (86.4%) and GPT-5.4 (82.8%). BenchLM does not use these results to rank models overall.
Claude Mythos Preview
Anthropic
Muse Spark
Meta
GPT-5.4
OpenAI
The published CharXiv snapshot is tightly clustered at the top: Claude Mythos Preview sits at 93.2%, while the third row is only 10.4 points behind. The broader top-10 spread is 32.3 points, so the benchmark still separates strong models even when the leaders cluster.
5 models have been evaluated on CharXiv. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. CharXiv is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
Year
2024
Tasks
Scientific chart reasoning
Format
Chart understanding and reasoning
Difficulty
Scientific visualization reasoning
CharXiv evaluates a model's ability to reason about real-world scientific charts rather than simple visual QA. With-tools and without-tools variants isolate raw visual reasoning from tool-augmented performance.
Version
CharXiv 2024
Refresh cadence
Annual
Staleness state
Refreshing
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
A scientific chart reasoning benchmark that tests whether models can understand, interpret, and reason about complex scientific visualizations including plots, diagrams, and data charts.
Claude Mythos Preview by Anthropic currently leads with a score of 93.2% on CharXiv.
5 AI models have been evaluated on CharXiv on BenchLM.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.