Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Multimodal & Grounded benchmark
CharXiv Reasoning without tools. Every model the catalog carries a published CharXiv w/o tools value for, ranked by that value.
Tool-free variant of CharXiv that isolates raw visual reasoning ability without code execution or tool augmentation. 19 models in the ModelScale catalog have a published value.
Each point is a model with a published CharXiv w/o tools value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
19 models with a published CharXiv w/o tools value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Chart understanding without tools |
|---|---|---|---|
| 1 | Moonshot AI | 89 | |
| 2 | Anthropic | 88.9 | |
| 3 | Alibaba | 88.4 |
What the benchmark measures and how its value is published, from the source’s own record.
CharXiv w/o tools (CharXiv Reasoning without tools) is a multimodal & grounded benchmark published by CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs. Tool-free variant of CharXiv that isolates raw visual reasoning ability without code execution or tool augmentation. Its evaluation set is described by the source as scientific chart reasoning (tool-free). The source rates its difficulty as scientific visualization reasoning.
Each model's result is published by CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs as chart understanding without tools. ModelScale prints that value exactly as published, and never converts or rescales it. This page ranks the 19 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
All leaderboards · Multimodal & Grounded capability leaderboard
| 4 | 86.2 |
| 5 | TMInkling 2 Small | Thinking Machines Lab | 85.6 |
| 6 | Moonshot AI | 84.8 |
| 7 | Alibaba | 84.6 |
| 8 | 84.5 |
| 9 | Alibaba | 83.7 |
| 10 | Alibaba | 83.5 |
| 11 | DSdots3-note Preview | Dots Studio | 83.1 |
| 12 | Anthropic | 82.1 |
| 12 | Anthropic | 82.1 |
| 14 | TMInkling 2 Mini | Thinking Machines Lab | 80.6 |
| 15 | Anthropic | 80.5 |
| 16 | TMInkling | Thinking Machines Lab | 78.1 |
| 17 | TMInkling-Small | Thinking Machines Lab | 77.4 |
| 18 | Anthropic | 77 |
| 19 | LiquidAI | 70.2 |
Evidence key: ObservedLast good
Rows are ordered by the value CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.