multimodal benchmark
ChartQA Leaderboard
ChartQA is a large-scale benchmark comprising 9.6K human-written questions and 23.1K questions generated from human-written chart summaries, designed to evaluate models' abilities in visual and logical reasoning over charts.
Updated Aug 17, 2026
Models26
Model coverage26
MetricScore
EvidenceB
ChartQA Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score90.8% | Percentile100.0% | Participants26 | EvidenceC | Evaluated |
| Rank02 | ModelME | Score90.0% | Percentile96.0% | Participants26 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score89.5% | Percentile92.0% | Participants26 | EvidenceC | Evaluated |
| Rank04 | ModelAM Nova ProAmazon | Score89.2% | Percentile88.0% | Participants26 | EvidenceC | Evaluated |
| Rank05 | ModelME | Score88.8% | Percentile84.0% | Participants26 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score88.3% | Percentile80.0% | Participants26 | EvidenceC | Evaluated |
| Rank07 | ModelMA | Score88.1% | Percentile76.0% | Participants26 | EvidenceC | Evaluated |
| Rank08 | ModelMA | Score87.4% | Percentile72.0% | Participants26 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score87.3% | Percentile68.0% | Participants26 | EvidenceC | Evaluated |
| Rank10 | ModelAMNova LiteAmazon | Score86.8% | Percentile64.0% | Participants26 | EvidenceC | Evaluated |
| Rank11 | ModelDE | Score86.0% | Percentile60.0% | Participants26 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score85.7% | Percentile56.0% | Participants26 | EvidenceC | Evaluated |
| Rank13 | ModelME | Score85.5% | Percentile52.0% | Participants26 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score85.3% | Percentile48.0% | Participants26 | EvidenceC | Evaluated |
| Rank15 | ModelDE | Score84.5% | Percentile44.0% | Participants26 | EvidenceC | Evaluated |
| Rank16 | ModelME | Score83.4% | Percentile40.0% | Participants26 | EvidenceC | Evaluated |
| Rank17 | ModelMI | Score81.8% | Percentile36.0% | Participants26 | EvidenceC | Evaluated |
| Rank18 | ModelMA | Score81.8% | Percentile32.0% | Participants26 | EvidenceC | Evaluated |
| Rank19 | ModelMI | Score81.4% | Percentile28.0% | Participants26 | EvidenceC | Evaluated |
| Rank20 | ModelLA | Score81.3% | Percentile24.0% | Participants26 | EvidenceC | Evaluated |
| Rank21 | ModelDE | Score81.0% | Percentile20.0% | Participants26 | EvidenceC | Evaluated |
| Rank22 | ModelCO | Score80.8% | Percentile16.0% | Participants26 | EvidenceC | Evaluated |
| Rank23 | ModelGO | Score78.0% | Percentile12.0% | Participants26 | EvidenceC | Evaluated |
| Rank24 | ModelXA | Score76.1% | Percentile8.0% | Participants26 | EvidenceC | Evaluated |
| Rank25 | ModelGO | Score75.7% | Percentile4.0% | Participants26 | EvidenceC | Evaluated |
| Rank26 | ModelGO | Score68.8% | Percentile0.0% | Participants26 | EvidenceC | Evaluated |
ChartQA Highlights
The leading models and scores on this benchmark.
ChartQA Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for ChartQA
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis chartqa AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ANClaude 3.5 SonnetAnthropic
- Score
- 90.8%
- Speed
- Up to 101 tok/s via Bedrock
Strengths
- Ranks #1 of 26 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures ChartQA, not total model capability
- 02MELlama 4 MaverickMeta
- Score
- 90.0%
- Speed
- Up to 639 tok/s via Sambanova
Strengths
- Ranks #2 of 26 compared models
- 96th percentile on this benchmark
- C evidence result
Considerations
- This result measures ChartQA, not total model capability
- 03ACQwen2.5 VL 72B InstructAlibaba Cloud / Qwen Team
- Score
- 89.5%
Strengths
- Ranks #3 of 26 compared models
- 92th percentile on this benchmark
- C evidence result
Considerations
- This result measures ChartQA, not total model capability
- 04AMNova ProAmazon
- Score
- 89.2%
- Price
- $0.80 input / $3.2 output per 1M tokens
- Speed
- Up to 100 tok/s via Bedrock
Strengths
- Ranks #4 of 26 compared models
- 88th percentile on this benchmark
- C evidence result
Considerations
- This result measures ChartQA, not total model capability
- 05MELlama 4 ScoutMeta
- Score
- 88.8%
- Speed
- Up to 776 tok/s via Groq
Strengths
- Ranks #5 of 26 compared models
- 84th percentile on this benchmark
- C evidence result
Considerations
- This result measures ChartQA, not total model capability
Selection summary
Best AI Models for ChartQA
Claude 3.5 Sonnet currently leads ChartQA with 90.8%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is ChartQA?
What ChartQA measures and how its scores work.
ChartQA is a large-scale benchmark comprising 9.6K human-written questions and 23.1K questions generated from human-written chart summaries, designed to evaluate models' abilities in visual and logical reasoning over charts.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- ChartQA
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- chartqa|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about ChartQA.
Which model scores highest on ChartQA?
Claude 3.5 Sonnet is currently ranked first with 90.8%.
What does ChartQA measure?
ChartQA is a large-scale benchmark comprising 9.6K human-written questions and 23.1K questions generated from human-written chart summaries, designed to evaluate models' abilities in visual and logical reasoning over charts.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
26 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.