# CharXiv Reasoning — AI model rankings

Can the AI read and reason about complex charts and figures, not just text? Higher is better.

16 tracked models have a published CharXiv Reasoning score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Muse Spark | Meta | 88.9% | Lab | Apr 8 2026 |
| 2 | Muse Spark 1.1 | Meta | 88.4% | Lab | Jul 9 2026 |
| 2 | Qwen3.8-Max | Qwen | 88.4% | Lab | Aug 3 2026 |
| 4 | Gemini 3.8 Flash | Google | 86.2% | Lab | Sep 2 2026 |
| 5 | Kimi K3 | Moonshot AI | 84.8% | Lab | Jul 16 2026 |
| 6 | Qwen3.8-Flash-Next | Qwen | 84.6% | Lab | Aug 26 2026 |
| 7 | Gemini 3.7 Flash | Google | 84.5% | Lab | Aug 13 2026 |
| 8 | Gemini 3.5 Flash | Google | 84.2% | Lab | May 19 2026 |
| 9 | GPT-5.5 | OpenAI | 84.1% | Lab | Apr 23 2026 |
| 10 | Gemini 3.1 Pro | Google | 83.3% | Lab | Feb 19 2026 |
| 11 | Claude Opus 4.7 | Anthropic | 82.1% | Lab | Apr 16 2026 |
| 12 | Qwen3.5 | Qwen | 80.8% | Lab | Feb 16 2026 |
| 13 | Gemini 3.0 Flash | Google | 80.3% | Lab | Dec 17 2025 |
| 14 | Muse Glimmer | Meta | 78.8% | Lab | Aug 10 2026 |
| 15 | Qwen3.6 | Qwen | 78% | Lab | Apr 16 2026 |
| 16 | Claude Sonnet 4.6 | Anthropic | 72.4% | Lab | Feb 17 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/charxiv-reasoning
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.