Vision Leaderboard
This page shows the top 30 models on the Arena AI leaderboard, snapshotted daily so you can see how recent models are doing at a glance. See the Arena AI site for the full leaderboard, filters and latest changes.
This leaderboard measures how multimodal models perform on image-understanding tasks.
Data updated: 2026-10-16 13:02:49 UTC / 2026-10-16 21:02:49 CST (Beijing time)
Top 30
| Rank | Rank Spread | Model | Score | Votes | Price $/M | Context |
|---|---|---|---|---|---|---|
| 1 | 1-10 | Claude Fable 5 (High) 1 | 1308 (±7) | 14,175 | $10/$50 | 1M |
| 2 | 1-19 | qwen3.8-max 1 | 1301 (±7) | 10,017 | $2/$6 | 1M |
| 3 | 1-19 | Claude Opus 4.6 (High) 1 | 1299 (±6) | 22,304 | $5/$25 | 1M |
| 4 | 1-22 | Claude Opus 4.7 1 | 1298 (±7) | 23,165 | $5/$25 | 1M |
| 5 | 1-22 | Claude Opus 4.7 (High) 1 | 1298 (±7) | 22,733 | $5/$25 | 1M |
| 6 | 1-26 | Gemini 3.7 Flash (High) 1 | 1297 (±10) Preliminary | 4,524 | $0.75/$3.75 | 1M |
| 7 | 2-25 | Claude Opus 4.6 1 | 1294 (±6) | 26,853 | $5/$25 | 1M |
| 8 | 1-37 | gpt-6.1-sol-max 1 | 1294 (±14) | 1,813 | $2/$10 | 1.1M |
| 9 | 1-31 | Muse Spark 1 | 1294 (±9) | 5,872 | N/A | N/A |
| 10 | 1-37 | Muse Spark 1.2 (xHigh) 1 | 1293 (±14) | 1,954 | $1.25/$4.25 | N/A |
| 11 | 1-36 | Claude Opus 5.5 (High) 1 | 1293 (±13) | 2,508 | $4/$20 | 1M |
| 12 | 2-36 | Muse Spark 1.3 (Max) 1 | 1290 (±10) | 4,550 | N/A | N/A |
| 13 | 2-34 | Gemini 3 Pro 1 | 1289 (±7) | 13,639 | $2/$12 | 1M |
| 14 | 2-37 | Gemini 3.8 Flash (High) 1 | 1289 (±11) Preliminary | 3,792 | $0.75/$3.75 | 1M |
| 15 | 2-35 | Claude Opus 5 (High) 1 | 1287 (±7) | 16,665 | $5/$25 | 1M |
| 16 | 2-37 | gpt-5.6-sol-xhigh 1 | 1287 (±7) | 10,907 | $4/$20 | N/A |
| 17 | 2-38 | Claude Fable 5.1 (Max) 1 | 1286 (±10) | 4,648 | $10/$50 | 1M |
| 18 | 2-38 | muse-spark-1.3 (xHigh) 1 | 1286 (±10) | 4,077 | N/A | N/A |
| 19 | 2-39 | gpt-6-astra-max 1 | 1286 (±11) | 3,590 | N/A | N/A |
| 20 | 4-36 | gpt-5.5 1 | 1286 (±6) | 27,993 | $5/$30 | 1.1M |
| 21 | 4-37 | claude-opus-4-8-high 1 | 1286 (±6) | 18,341 | $5/$25 | 1M |
| 22 | 4-37 | gemini-3.5-flash-medium 1 | 1285 (±7) | 13,180 | $0.75/$4.50 | 1M |
| 23 | 6-37 | gpt-5.4-high 1 | 1285 (±6) | 29,881 | $2.50/$15 | 1.1M |
| 24 | 6-38 | gemini-3.5-flash-high 1 | 1284 (±7) | 13,103 | $0.75/$4.50 | 1M |
| 25 | 8-39 | gpt-5.5-high 1 | 1281 (±6) | 26,281 | $5/$30 | 1.1M |
| 26 | 6-40 | gemini-3.6-flash-high 1 | 1281 (±8) | 8,127 | $0.75/$3.75 | 1M |
| 27 | 7-40 | grok-4.5 1 | 1280 (±7) | 11,764 | $2/$6 | 500K |
| 28 | 8-40 | Muse Spark 1.1 1 | 1280 (±7) | 11,994 | $1.25/$4.25 | 1M |
| 29 | 8-39 | Gemini 3.1 Pro Preview 1 | 1280 (±5) | 46,320 | $1/$6 | 1M |
| 30 | 8-40 | claude-opus-4-8 1 | 1279 (±7) | 18,852 | $5/$25 | 1M |
How to read this table
- Rank / Rank Spread: relative position estimated from battle votes. The spread shows how far the rank moves within the confidence interval.
- Score: a relative score, useful for comparing the models listed at the same moment.
- Votes / Sessions: sample size. Rankings move more when there are fewer votes.
- Price $/M: reference price per million tokens for input / output.
- Context: maximum context length the model supports.
Three things to check before choosing a model
- Whether the provider actually serves the model, and whether it is available in your region and for your account;
- Whether the API price, rate limits and context window fit your workload;
- Run a small test on three to five real tasks instead of relying on the overall ranking alone.
Data source
Data comes from the Arena AI Vision Leaderboard and is refreshed daily by GitHub Actions. For model pricing, licenses and capabilities, check the model provider's own documentation.