Skip to main content
Benchmark Registry
ModelsBenchmarksOrganizationsCompare

CharXiv

Benchmark
CharXiv
Evaluated by
Google DeepMind
Release date
July 21, 2026
Version
CharXiv Reasoning — Search and code execution
Metric
CharXiv Accuracy

Results

3 results

LatestHistory
All providersGoogle (DeepMind)
CharXiv Reasoning — Search and code execution results
ProviderModelScoreSourceRegistry No.
Google (DeepMind)Gemini 3.7 Flash88.7%Source for Gemini 3.7 Flash on CharXiv Reasoning — Search and code execution (opens in a new tab)30010
Google (DeepMind)Gemini 3.5 Flash84.9%Source for Gemini 3.5 Flash on CharXiv Reasoning — Search and code execution (opens in a new tab)30006
Google (DeepMind)Gemini 3.6 Flash89.4%Source for Gemini 3.6 Flash on CharXiv Reasoning — Search and code execution (opens in a new tab)30007
Page 1 of 1

© 2026 Densa Labs

Toggle between the data update date and the application build time.
Legal
Color theme