ChartFI - Knowledge Integration: leaderboard

ChartFI (2026) tests how faithfully and insightfully multimodal LLMs describe complex scientific charts, over 896 charts from peer-reviewed arXiv papers with author-grounded reference descriptions. This board is the knowledge integration efficiency part of the Acuity score, rated by Gemini-3.1-Pro on a 1 to 5 Likert scale and averaged over charts; higher is better.

Metric: Acuity sub-dimension knowledge integration efficiency: mean Gemini-3.1-Pro rating on a 5-point Likert scale (1-5), descriptions of the 896 ChartFI-Bench scientific charts (arXiv papers accepted at peer-reviewed venues), temperature 0 and top-p 1; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-5.43.31#135
2Qwen 3.5 Plus3.18#242
3Gemini 3 Flash3.17#175
4Qwen 3.5 27B2.99#375
5Claude Sonnet 4.62.96#173
6Llama 4 Scout2.23#1818
7InternVL3.5-14B1.91#1504

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=chartfi-knowledge-integration · How It Works · Data refreshed daily, snapshot 2026-11-21.