Chart Arena
We show this table for reference; we do not rank on it.
Kaggle's Chart Arena has models turn a small dataset and a brief into one chart, and people vote for the better chart.
Evaluation results published by Kaggle. How Kaggle Benchmarks reruns tests.
Arena rating on Chart Arena v1 — Latest evaluation November 12, 2026
We mirror the published arena rating view for Chart Arena v1. GPT-6.1 Sol has the highest reported value at 1627. A rating gap smaller than the two rows' intervals is not a reliable difference. We do not use these results to rank models overall.
GPT-6.1 Sol
42 votes; ±136; Evaluated 2026-11-12
Claude Opus 5.5
41 votes; ±136; Evaluated 2026-11-12
GPT-6 Astra
49 votes; ±128; Evaluated 2026-11-12
15 source rowsCross-domainCurrentDisplay onlyUpdated Latest evaluation November 12, 2026
Arena rating table (15 source rows)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | GPT-6.1 SolCreator not reported42 votes; ±136; Evaluated 2026-11-12 | 1627 | Not reported | Unknown |
| 2 | Claude Opus 5.5Creator not reported41 votes; ±136; Evaluated 2026-11-12 | 1619 | Not reported | Unknown |
| 3 | GPT-6 AstraCreator not reported49 votes; ±128; Evaluated 2026-11-12 | 1606 | Not reported | Unknown |
| 4 | Claude Fable 5.1Creator not reported39 votes; ±136; Evaluated 2026-11-12 | 1598 | Not reported | Unknown |
| 5 | Gemini 3.5 Flash-LiteCreator not reported44 votes; ±131; Evaluated 2026-11-12 | 1558 | Not reported | Unknown |
| 6 | Gemini 4 ArgonCreator not reported53 votes; ±124; Evaluated 2026-11-12 | 1521 | Not reported | Unknown |
| 7 | Gemini 4 CarbonCreator not reported49 votes; ±126; Evaluated 2026-11-12 | 1506 | Not reported | Unknown |
| 8 | Grok 4.7Creator not reported44 votes; ±129; Evaluated 2026-11-12 | 1504 | Not reported | Unknown |
| 9 | Gemini 3.8 FlashCreator not reported56 votes; ±123; Evaluated 2026-11-12 | 1472 | Not reported | Unknown |
| 10 | GLM-5.3-FlashCreator not reported45 votes; ±129; Evaluated 2026-11-12 | 1469 | Not reported | Unknown |
| 11 | Grok 4.5Creator not reported50 votes; ±127; Evaluated 2026-11-12 | 1467 | Not reported | Unknown |
| 12 | Claude Sonnet 5.5Creator not reported50 votes; ±126; Evaluated 2026-11-12 | 1456 | Not reported | Unknown |
| 13 | GPT-6 LunaCreator not reported37 votes; ±137; Evaluated 2026-11-12 | 1409 | Not reported | Unknown |
| 14 | GPT-5.6 LunaCreator not reported46 votes; ±133; Evaluated 2026-11-12 | 1365 | Not reported | Unknown |
| 15 | Gemma 4 26B A4BCreator not reported49 votes; ±133; Evaluated 2026-11-12 | 1322 | Not reported | Unknown |
How to read this leaderboard
Read the rating as a preference ranking for charts made from a dataset and a brief among the models on this board, with its vote count and interval. It does not measure accuracy.
Operator receipt: 15 sourced rows are currently displayable on this page; the highest published score among these 15 source rows is GPT-6.1 Sol at 1627, which does not establish a market leader.
Honest limit: The rating moves as votes arrive, and Kaggle publishes no confidence level for its interval. Rows with few votes have wide intervals, so neighbouring models are often not separable. Voters are self-selected Kaggle visitors. The board declares no license; we keep published aggregates with attribution and no task or output contents.
How to read Chart Arena
Briefs range from time series and distributions to funnels and forecast uncertainty. Each model returns one self-contained SVG chart, and all models run in the same agent harness (mini-swe-agent). Kaggle shows two anonymous outputs for the same prompt and a person picks the better one. The Arena column is the relative rating Kaggle fits to those votes, published with an interval; it is not a percentage.
The rating moves as votes arrive, and Kaggle publishes no confidence level for its interval. Rows with few votes have wide intervals, so neighbouring models are often not separable. Voters are self-selected Kaggle visitors. The board declares no license; we keep published aggregates with attribution and no task or output contents.
Snapshot
A rating gap smaller than the two rows' intervals is not a reliable difference. The table retains each source configuration and its reported value.
15 source rows are shown for Chart Arena. The benchmark falls in the Cross-domain category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. Chart Arena is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Chart Arena
Year
2026
Tasks
13 chart briefs with small datasets, rated by human votes
Format
Relative rating fitted to human pairwise votes
Difficulty
Data visualisation
Briefs range from time series and distributions to funnels and forecast uncertainty. Each model returns one self-contained SVG chart, and all models run in the same agent harness (mini-swe-agent). Kaggle shows two anonymous outputs for the same prompt and a person picks the better one. The Arena column is the relative rating Kaggle fits to those votes, published with an interval; it is not a percentage. These results are display-only and do not enter overall or category rankings.
Freshness and provenance
Version
Chart Arena 2026
Refresh cadence
Quarterly
Staleness state
Current
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 Chart Arena measure?
Kaggle's Chart Arena has models turn a small dataset and a brief into one chart, and people vote for the better chart.
Which source row has the highest reported Chart Arena value?
GPT-6.1 Sol has the highest reported Chart Arena value at 1627. A rating gap smaller than the two rows' intervals is not a reliable difference. BenchLM shows this benchmark for display only and does not use it in overall rankings.
How many models are evaluated on Chart Arena?
The Latest evaluation November 12, 2026 snapshot contains 15 AI models.
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.