OverviewAll LeaderboardsMethodology
New Chat

Vision Arena

What is the best model for vision tasks?

Nov 20, 2026
1,606,278 votes
168 models

Vision Arena

Frontier
AnthropicClaude Fable 5 (High)1,309 rating. 4 points above rank 2.
Daily Driver
Gemini 3.7 Flash (High)Near-frontier performance at 1/10th the price.
Efficient
Gemma 4 31BSolid performance at 1/100th the price.
1,2001,300
  1. 1,309
    Anthropic
    Claude Fable 5High
  2. 1,305
    Anthropic
    Claude Fable 5.2Max
  3. 1,299
    Anthropic
    Claude Opus 4.7
  4. 1,298
    Anthropic
    Claude Opus 4.6High
  5. 1,298
    Anthropic
    Claude Opus 4.7High
  6. 1,294
    Gemini 3.7 FlashHigh
  7. 1,294
    Meta
    Muse Spark
  8. 1,294
    Qwen3.8-Max
  9. 1,293
    Anthropic
    Claude Opus 5.5High
  10. 1,293
    Meta
    Muse Spark 1.2xHigh
  11. 1,293
    Anthropic
    Claude Opus 4.6
  12. 1,293
    Gemini 3.8 FlashHigh
  13. 1,290
    GPT-6.1 SolMax
  14. 1,290
    Anthropic
    Claude Fable 5.1Max
  15. 1,289
    Gemini 3 Pro
Model RankingScoreCost per task (P50)Blended price $/M (3:1)
  1. 1AnthropicClaude Fable 5 (High)1,309$0.21$40
  2. 2AnthropicClaude Fable 5.2 (Max)1,305$0.39$32
  3. 3AnthropicClaude Opus 4.71,299$0.09$20
  4. 4AnthropicClaude Opus 4.6 (High)1,298$0.10$20
  5. 5AnthropicClaude Opus 4.7 (High)1,298$0.10$20
  6. 6Gemini 3.7 Flash (High)1,294$0.02$3
  7. 7MetaMuse Spark1,294N/AN/A
  8. 8Qwen3.8-Max1,294$0.04$5
  9. 9AnthropicClaude Opus 5.5 (High)1,293$0.10$16
  10. 10MetaMuse Spark 1.2 (xHigh)1,293$0.03$3.50
  11. 11AnthropicClaude Opus 4.61,293$0.09$20
  12. 12Gemini 3.8 Flash (High)1,293$0.02$3
  13. 13GPT-6.1 Sol (Max)1,290$0.06$8
  14. 14AnthropicClaude Fable 5.1 (Max)1,290$0.44$40
  15. 15Gemini 3 Pro1,289N/A$9.50

Pareto frontier

Best score for the cost
1,2001,2501,300
AnthropicAnthropic
$0.001$0.01$0.10

Median cost per task

Frontier
AnthropicClaude Fable 5 (High)1,309 rating. 4 points above rank 2.
Daily Driver
Gemini 3.7 Flash (High)Near-frontier performance at 1/10th the price.
Efficient
Gemma 4 31BSolid performance at 1/100th the price.
On the frontierScore$/task
  1. AnthropicClaude Fable 5 (High)1,309$0.21
  2. AnthropicClaude Opus 4.71,299$0.09
  3. Gemini 3.7 Flash (High)1,294$0.02
  4. Gemini 3.8 Flash (High)1,293$0.02
  5. Grok 4.51,280$0.01
  6. GLM-5.3-Flash1,274$0.01
  7. Gemma 4 31B1,265$0.01
96 models not plotted

These models are on the leaderboard but do not appear on the Pareto chart because they have insufficient cost data. The numbers beside their names show their leaderboard ranks.

  • 7Muse Spark
  • 15Gemini 3 Pro
  • 37Gemini 3 Flash
  • 50Dola Seed 2.0 Pro
  • 54GPT-5.1 (High)
  • 59Gemini 2.5 Pro
  • 64GPT-5.2 (High)
  • 67GPT-5.1
  • 68ChatGPT-4o Latest (20250326)
  • 69Kimi K2.5 (Instant)
  • 75Gemini 2.5 Flash Preview (09-2025)
  • 76GPT-5.2
  • 78GPT-5 Chat
  • 79GPT-4.5 Preview (2025-02-27)
  • 80ERNIE-5.0-Preview (1220)
  • 82o3 (2025-04-16)
  • 83Gemini 2.5 Flash
  • 85Qwen3-VL-235B-A22B-Instruct
  • 86GPT-4.1 (2025-04-14)
  • 87GPT-5 (High)
  • 88NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)
  • 90Claude Opus 4 (Thinking 16K, 20250514)
  • 91Claude Sonnet 4 (Thinking 32K, 20250514)
  • 92o4-mini (2025-04-16)
  • 93GPT-4.1 mini (2025-04-14)
  • 98Claude 3.7 Sonnet (Thinking 32K, 20250219)
  • 99o1 (2024-12-17)
  • 100Claude Opus 4 (20250514)
  • 101Gemini 2.5 Flash-Lite Preview (Thinking, 06-17)
  • 102Claude Sonnet 4 (20250514)
  • 103Qwen3-VL-235B-A22B-Thinking
  • 104GPT-5 mini (High)
  • 105Qwen-VL-Max (2025-08-13)
  • 106Grok 4 (0709)
  • 107Gemini 1.5 Pro (002)
  • 108Claude 3.7 Sonnet (20250219)
  • 109Gemini 2.5 Flash-Lite Preview (Non-Thinking, 09-2025)
  • 110Gemini 2.0 Flash (001)
  • 111GPT-4o (2024-05-13)
  • 112Claude 3.5 Sonnet (20241022)
  • 113GLM-4.6V
  • 114Gemma 3 27B IT
  • 115Mistral Medium 3.1 (2508)
  • 116Mistral Medium 3 (2505)
  • 117Step-1o Turbo (202506)
  • 118Hunyuan Vision 1.5 Thinking
  • 119GLM-4.5V
  • 120Hunyuan Large Vision
  • 121GPT-5 nano (High)
  • 122Claude 3.5 Sonnet (20240620)
  • 123Llama 4 Maverick 17B 128E Instruct
  • 124Step-3
  • 125Gemini 1.5 Flash (002)
  • 126Mistral Small 3.2 (2506)
  • 127Gemini 2.0 Flash-Lite Preview (02-05)
  • 128Llama 4 Scout 17B 16E Instruct
  • 129Claude 3.5 Haiku (20241022)
  • 130Mistral Small 3.1 24B Instruct (2503)
  • 131Step-1o Vision 32K (High Res)
  • 132Qwen2.5-VL-72B-Instruct
  • 133Gemini 1.5 Pro (001)
  • 134GPT-4o (2024-08-06)
  • 135Qwen2.5-VL-32B-Instruct
  • 136Molmo 2 8B
  • 137GPT-4 Turbo (2024-04-09)
  • 138GPT-4o mini (2024-07-18)
  • 139Pixtral Large (2411)
  • 140GPT-4.1 nano (2025-04-14)
  • 141Qwen2-VL-72B
  • 142Qwen-VL-Max (1119)
  • 143Gemini 1.5 Flash-8B (001)
  • 144Step-1V 32K
  • 145Claude 3 Opus (20240229)
  • 146Gemini 1.5 Flash (001)
  • 147Molmo 72B (0924)
  • 148Hunyuan Standard Vision (2024-12-31)
  • 149Llama 3.2 90B Vision Instruct
  • 150Qwen2-VL-7B-Instruct
  • 151Pixtral 12B (2409)
  • 152InternVL2 26B
  • 153Amazon Nova Lite 1.0
  • 154Amazon Nova Pro 1.0
  • 155Claude 3 Sonnet (20240229)
  • 156Yi Vision
  • 157Aya Vision 32B
  • 158Claude 3 Haiku (20240307)
  • 159Molmo 7B-D (0924)
  • 160Llama 3.2 11B Vision Instruct
  • 161nvila-internal-15b-v1
  • 162LLaVA-OneVision Qwen2 72B OV
  • 163LLaVA 1.6 34B
  • 164CogVLM2 Llama3 Chat 19B
  • 165MiniCPM-V 2.6
  • 166InternVL2 4B
  • 167Phi-3.5 Vision Instruct
  • 168Phi-3 Vision 128K Instruct

Default Leaderboard Plots

1 model with no battles against the other models is hidden.

Previous
Search

Which model is the best at online search?

Next
Document

Which model is the best at analyzing documents?

USE CASES

  • Chat with AI
  • Build Apps & Websites
  • Write & Edit Text
  • Search the Web
  • Generate Images
  • Generate Videos
  • Chose any model
  • Compare Models Side by Side

LEADERBOARD RANKINGS

  • Overall
  • Agent
  • Alignment Index
  • Text
  • WebDev
  • Image-to-WebDev
  • Text to Image
  • Image Edit
  • Text to Video
  • Image to Video
  • Video Edit
  • Vision
  • Document
  • Search

COMPANY

  • About Us
  • How It Works
  • Blog
  • Careers
  • Leaderboard Changelog
  • Product Changelog
  • Help Center
  • FAQ

LEGAL

  • Terms
  • Privacy
  • Cookies

FOLLOW

  • X
  • LinkedIn
  • YouTube
  • Discord

© Arena Intelligence