benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
CharXiv With Tools leaderboard
CharXiv With Tools
4 models tested · Updated 2026-07-21 · Verified sources only
Gemini 3.6 Flash
leads at
89.4%
1
Gemini 3.6 Flash
Google ·
Blog/Google DeepMind
· 2026-07-21
With tools. Up from 84.9% (Gemini 3.5 Flash) and 83.2% (Gemini 3.1 Pro).
89.4%
2
Claude Sonnet 5
Anthropic ·
Blog/Google DeepMind (cross-ref)
· 2026-07-21
Google's measurement. Near Gemini 3.6 Flash (89.4).
88.3%
3
Gemini 3.5 Flash
Google DeepMind ·
Blog/Google DeepMind (cross-ref)
· 2026-07-21
Google official. CharXiv with tools variant.
84.9%
4
Gemini 3.1 Pro
Google DeepMind ·
Blog/Google DeepMind (cross-ref)
· 2026-07-21
Google official CharXiv with tools.
83.2%