multimodal benchmark
CharXiv-R
CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements. It evaluates multimodal large language models on their ability to understand and reason about scientific charts from arXiv papers through various reasoning tasks.
Updated Aug 7, 2026
Published models47
Registry coverage47
MetricScore
EvidenceB
CharXiv-R leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
Score distribution
Top published rows on the benchmark's original scale.
CharXiv-R highlights
The top published results on this benchmark's own scale.
What is CharXiv-R?
Definition and scoring fields from the benchmark registry.
CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements. It evaluates multimodal large language models on their ability to understand and reason about scientific charts from arXiv papers through various reasoning tasks.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- CharXiv-R
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- charxiv-r|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about CharXiv-R.
Which model scores highest on CharXiv-R?
Claude Mythos Preview is currently ranked first with 93.2%.
What does CharXiv-R measure?
CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements. It evaluates multimodal large language models on their ability to understand and reason about scientific charts from arXiv papers through various reasoning tasks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
47 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.