CharXiv Reasoning without tools (CharXiv w/o tools)
Tool-free variant of CharXiv that isolates raw visual reasoning ability without code execution or tool augmentation.
Top models on CharXiv w/o tools — November 13, 2026
As of November 13, 2026, Kimi K3.1 Pro leads the CharXiv w/o tools leaderboard with 89% , followed by Claude Mythos 5 (88.9%) and Qwen3.8 Max (88.4%).
Kimi K3.1 Pro
Moonshot AI
Reported by Moonshot AIClaude Mythos 5
Anthropic
Reported by AnthropicQwen3.8 Max
Alibaba
Reported by Alibaba17 modelsMultimodal & Grounded7% of Multimodal reference weightRefreshingUpdated November 13, 2026
Leaderboard (17 models)
All 17 results on this page were reported by the labs that built the models.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Kimi K3.1 ProMoonshot AIReported by Moonshot AI | 89% | Reported by Moonshot AI | Not reported | Open |
| 2 | Claude Mythos 5AnthropicReported by Anthropic | 88.9% | Reported by Anthropic | Not reported | Closed |
| 3 | Qwen3.8 MaxAlibabaReported by Alibaba | 88.4% | Reported by Alibaba | Not reported | Open |
| 4 | Gemini 3.8 FlashGoogleReported by Google | 86.2% | Reported by Google | Not reported | Closed |
| 5 | Inkling 2 SmallThinking Machines LabReported by Thinking Machines Lab | 85.6% | Reported by Thinking Machines Lab | Not reported | Closed |
| 6 | Kimi K3Moonshot AIReported by Moonshot AI | 84.8% | Reported by Moonshot AI | Not reported | Open |
| 7 | Qwen3.8-Flash-NextAlibabaReported by Alibaba | 84.6% | Reported by Alibaba | Not reported | Open |
| 8 | Gemini 3.7 FlashGoogleReported by Google | 84.5% | Reported by Google | Not reported | Closed |
| 9 | Qwen3.8-27BAlibabaReported by Alibaba | 83.7% | Reported by Alibaba | Not reported | Open |
| 10 | Qwen3.8-Omni-FlashAlibabaReported by Alibaba | 83.5% | Reported by Alibaba | Not reported | Closed |
| 11 | dots3-note PreviewDots StudioReported by Dots Studio | 83.1% | Reported by Dots Studio | Not reported | Open |
| 12 | Claude Opus 4.7 (Adaptive)AnthropicReported by Anthropic | 82.1% | Reported by Anthropic | Not reported | Closed |
| 13 | Claude Opus 4.8AnthropicReported by Anthropic | 80.5% | Reported by Anthropic | Not reported | Closed |
| 14 | InklingThinking Machines LabReported by Thinking Machines Lab | 78.1% | Reported by Thinking Machines Lab | Not reported | Open |
| 15 | Inkling-SmallThinking Machines LabReported by Thinking Machines Lab | 77.4% | Reported by Thinking Machines Lab | Not reported | Open |
| 16 | Claude Sonnet 5AnthropicReported by Anthropic | 77% | Reported by Anthropic | Not reported | Closed |
| 17 | LFM3-VL-2BLiquidAIReported by LiquidAI | 70.2% | Reported by LiquidAI | Not reported | Open |
- Lab-reported
- Benchmark-run
- Independent
According to BenchLM.ai, Kimi K3.1 Pro leads the CharXiv w/o tools benchmark with a score of 89%, followed by Claude Mythos 5 (88.9%) and Qwen3.8 Max (88.4%). The top models are clustered within 0.6 points, suggesting this benchmark is nearing saturation for frontier models.
17 models have been evaluated on CharXiv w/o tools. The benchmark falls in the Multimodal & Grounded category. BenchAlign v5.8 gives CharXiv w/o tools 7% of the Multimodal reference weight, so it moves the Multimodal leaderboard. Reference weights are relative weights in the calibrated model, not fixed shares of a score.
About CharXiv w/o tools
Year
2024
Tasks
Scientific chart reasoning (tool-free)
Format
Chart understanding without tools
Difficulty
Scientific visualization reasoning
The tool-free CharXiv variant measures pure multimodal reasoning. Mythos Preview scores 86.1% without tools vs 93.2% with tools, demonstrating strong baseline chart reasoning.
Freshness and provenance
Version
CharXiv w/o tools 2024
Refresh cadence
Annual
Staleness state
Refreshing
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does CharXiv w/o tools measure?
Tool-free variant of CharXiv that isolates raw visual reasoning ability without code execution or tool augmentation.
Which model scores highest on CharXiv w/o tools?
Kimi K3.1 Pro by Moonshot AI currently leads with a score of 89% on CharXiv w/o tools.
How many models are evaluated on CharXiv w/o tools?
17 AI models have published results on CharXiv w/o tools in the BenchLM catalog.
Compare top models on CharXiv w/o tools
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.