Benchmark Almanac

Image-to-WebDev Arena

Claude Opus 5.5 1,753 leads since Sep 2026 Not in model cards

Users upload a screenshot, wireframe or mockup, two anonymous models build the website in Code Arena, and users vote for the better result. Votes become ratings with bootstrapped confidence intervals.

On 2026-10-19 the board listed 65 models and 191,568 votes, led by Claude Opus 5.5 (Max). It is a separate board from the text-prompt Code Arena WebDev board.

Made by Arena (formerly LMArena) Released Apr 2026 Web and app front ends Tasks kept private Last model added Oct 2026 Scores refreshed 2 Nov 2026
Level
Everyday Most adults, without training
Scored
Ranked by votes People pick the better of two answers, as in an arena, and the votes are turned into ratings.
Answers
Open-ended Free-form text, documents, designs or media judged as a whole.
Leaderboard
55
Models scored on its own leaderboard

Claude Opus 5.5 leads with 1,753

Best score per model from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

55 models. Scroll sideways for the rest. Bars start at 1,200.

  1. 1,753 Claude Opus 5.5
  2. 1,737 Claude Sonnet 5.5
  3. 1,726 GPT-6 Astra
  4. 1,722 Claude Fable 5.1
  5. 1,702 GPT-6.1 Sol
  6. 1,662 Claude Opus 5
  7. 1,660 GPT-6 Sol
  8. 1,640 Muse Spark 1.3
  9. 1,637 Qwen3.8 Max (0902)
  10. 1,623 Claude Fable 5
  11. 1,615 Qwen 3.8 Max
  12. 1,609 GPT-5.6 Sol
  13. 1,606 DeepSeek V4.1 Flash
  14. 1,603 Grok 4.6
  15. 1,599 Gemini 3.7 Flash
  16. 1,598 MiMo-V2.6 Pro
  17. 1,596 MiMo-V2.6 Flash
  18. 1,593 Gemini 3.8 Flash
  19. 1,586 Grok 4.7
  20. 1,579 Kimi K3
  21. 1,578 GLM-5.3-Flash
  22. 1,577 Claude Opus 4.7
  23. 1,577 Qwen 3.8 27B
  24. 1,571 GPT-6 Luna
  25. 1,562 Grok 4.5
  26. 1,550 Step 5 Preview
  27. 1,545 Gemini 3.6 Flash
  28. 1,543 Claude Opus 4.6
  29. 1,541 GPT-5.6 Terra
  30. 1,540 Claude Sonnet 5
  31. 1,540 Muse Spark 1.1
  32. 1,532 GPT-5.5
  33. 1,529 Claude Sonnet 4.6
  34. 1,521 Kimi K2.6
  35. 1,517 GPT-5.6 Luna
  36. 1,513 Kimi K2.7 Code
  37. 1,510 Seed2.1 Pro
  38. 1,504 Gemini 3.5 Flash
  39. 1,481 Gemini 3.1 Pro
  40. 1,472 Qwen 3.6 Plus
  41. 1,470 MiniMax-M3
  42. 1,450 Gemini 3 Flash
  43. 1,448 Gemini 3.5 Flash-Lite
  44. 1,443 GPT-5.3 Codex
  45. 1,441 Gemini 3 Pro
  46. 1,439 Kimi K2.5
  47. 1,436 GPT-5.4
  48. 1,425 GLM-5V-Turbo
  49. 1,421 GPT-5.1
  50. 1,410 Inkling-Small
  51. 1,407 Nemotron 3.5 Super VL
  52. 1,375 Grok 4.3
  53. 1,361 Mistral Large 3
  54. 1,336 Gemini 3.1 Flash-Lite
  55. 1,285 Gemini 2.5 Pro

Bars start at 1,200.

  1. 1 Claude Opus 5.5 max 1,753
  2. 2 Claude Sonnet 5.5 xhigh 1,737
  3. 3 GPT-6 Astra max 1,726
  4. 4 Claude Fable 5.1 max 1,722
  5. 5 GPT-6.1 Sol max 1,702
  6. 6 Claude Opus 5 max 1,662
  7. 7 GPT-6 Sol max 1,660
  8. 8 Muse Spark 1.3 max 1,640
  9. 9 Qwen3.8 Max (0902) 1,637
  10. 10 Claude Fable 5 high 1,623
  11. 11 Qwen 3.8 Max 1,615
  12. 12 GPT-5.6 Sol 1,609
  13. 13 DeepSeek V4.1 Flash max 1,606
  14. 14 Grok 4.6 high 1,603
  15. 15 Gemini 3.7 Flash high 1,599
  16. 16 MiMo-V2.6 Pro 1,598
  17. 17 MiMo-V2.6 Flash 1,596
  18. 18 Gemini 3.8 Flash high 1,593
  19. 19 Grok 4.7 xhigh 1,586
  20. 20 Kimi K3 max 1,579
  21. 21 GLM-5.3-Flash 1,578
  22. 22 Claude Opus 4.7 high 1,577
  23. 23 Qwen 3.8 27B 1,577
  24. 24 GPT-6 Luna max 1,571
  25. 25 Grok 4.5 1,562
  26. 26 Step 5 Preview high 1,550
  27. 27 Gemini 3.6 Flash high 1,545
  28. 28 Claude Opus 4.6 high 1,543
  29. 29 GPT-5.6 Terra 1,541
  30. 30 Claude Sonnet 5 high 1,540
  31. 31 Muse Spark 1.1 1,540
  32. 32 GPT-5.5 1,532
  33. 33 Claude Sonnet 4.6 1,529
  34. 34 Kimi K2.6 1,521
  35. 35 GPT-5.6 Luna 1,517
  36. 36 Kimi K2.7 Code 1,513
  37. 37 Seed2.1 Pro 1,510
  38. 38 Gemini 3.5 Flash medium 1,504
  39. 39 Gemini 3.1 Pro 1,481
  40. 40 Qwen 3.6 Plus 1,472
  41. 41 MiniMax-M3 1,470
  42. 42 Gemini 3 Flash 1,450
  43. 43 Gemini 3.5 Flash-Lite 1,448
  44. 44 GPT-5.3 Codex 1,443
  45. 45 Gemini 3 Pro 1,441
  46. 46 Kimi K2.5 thinking 1,439
  47. 47 GPT-5.4 1,436
  48. 48 GLM-5V-Turbo 1,425
  49. 49 GPT-5.1 high 1,421
  50. 50 Inkling-Small 1,410
  51. 51 Nemotron 3.5 Super VL nvfp4 1,407
  52. 52 Grok 4.3 1,375
  53. 53 Mistral Large 3 1,361
  54. 54 Gemini 3.1 Flash-Lite 1,336
  55. 55 Gemini 2.5 Pro Jun 2025 1,285
#ModelScoreSettingDateSource
1 Claude Opus 5.5 1,753 max Sep 2026 Official leaderboard
2 Claude Sonnet 5.5 1,737 xhigh Sep 2026 Official leaderboard
3 GPT-6 Astra 1,726 max Sep 2026 Official leaderboard
4 Claude Fable 5.1 1,722 max Sep 2026 Official leaderboard
5 GPT-6.1 Sol 1,702 max Sep 2026 Official leaderboard
6 Claude Opus 5 1,662 max Jul 2026 Official leaderboard
7 GPT-6 Sol 1,660 max Sep 2026 Official leaderboard
8 Muse Spark 1.3 1,640 max Sep 2026 Official leaderboard
9 Qwen3.8 Max (0902) 1,637 Sep 2026 Official leaderboard
10 Claude Fable 5 1,623 high Jun 2026 Official leaderboard
11 Qwen 3.8 Max 1,615 Aug 2026 Official leaderboard
12 GPT-5.6 Sol 1,609 xHigh, Codex Jul 2026 Official leaderboard
13 DeepSeek V4.1 Flash 1,606 max Sep 2026 Official leaderboard
14 Grok 4.6 1,603 high Aug 2026 Official leaderboard
15 Gemini 3.7 Flash 1,599 high Aug 2026 Official leaderboard
16 MiMo-V2.6 Pro 1,598 Sep 2026 Official leaderboard
17 MiMo-V2.6 Flash 1,596 Sep 2026 Official leaderboard
18 Gemini 3.8 Flash 1,593 high Sep 2026 Official leaderboard
19 Grok 4.7 1,586 xhigh Sep 2026 Official leaderboard
20 Kimi K3 1,579 max Jul 2026 Official leaderboard
21 GLM-5.3-Flash 1,578 Aug 2026 Official leaderboard
22 Claude Opus 4.7 1,577 high Apr 2026 Official leaderboard
23 Qwen 3.8 27B 1,577 Aug 2026 Official leaderboard
24 GPT-6 Luna 1,571 max Sep 2026 Official leaderboard
25 Grok 4.5 1,562 Jul 2026 Official leaderboard
26 Step 5 Preview 1,550 high Sep 2026 Official leaderboard
27 Gemini 3.6 Flash 1,545 high Jul 2026 Official leaderboard
28 Claude Opus 4.6 1,543 high Feb 2026 Official leaderboard
29 GPT-5.6 Terra 1,541 xHigh, Codex Jul 2026 Official leaderboard
30 Claude Sonnet 5 1,540 high Jun 2026 Official leaderboard
31 Muse Spark 1.1 1,540 Jul 2026 Official leaderboard
32 GPT-5.5 1,532 xHigh, Codex Apr 2026 Official leaderboard
33 Claude Sonnet 4.6 1,529 Feb 2026 Official leaderboard
34 Kimi K2.6 1,521 Apr 2026 Official leaderboard
35 GPT-5.6 Luna 1,517 xHigh, Codex Jul 2026 Official leaderboard
36 Kimi K2.7 Code 1,513 Jun 2026 Official leaderboard
37 Seed2.1 Pro 1,510 Jun 2026 Official leaderboard
38 Gemini 3.5 Flash 1,504 medium May 2026 Official leaderboard
39 Gemini 3.1 Pro 1,481 Feb 2026 Official leaderboard
40 Qwen 3.6 Plus 1,472 Mar 2026 Official leaderboard
41 MiniMax-M3 1,470 Jun 2026 Official leaderboard
42 Gemini 3 Flash 1,450 Dec 2025 Official leaderboard
43 Gemini 3.5 Flash-Lite 1,448 Jul 2026 Official leaderboard
44 GPT-5.3 Codex 1,443 Codex Feb 2026 Official leaderboard
45 Gemini 3 Pro 1,441 Nov 2025 Official leaderboard
46 Kimi K2.5 1,439 thinking Jan 2026 Official leaderboard
47 GPT-5.4 1,436 Mar 2026 Official leaderboard
48 GLM-5V-Turbo 1,425 Apr 2026 Official leaderboard
49 GPT-5.1 1,421 high Nov 2025 Official leaderboard
50 Inkling-Small 1,410 Jul 2026 Official leaderboard
51 Nemotron 3.5 Super VL 1,407 nvfp4 Oct 2026 Official leaderboard
52 Grok 4.3 1,375 Apr 2026 Official leaderboard
53 Mistral Large 3 1,361 Dec 2025 Official leaderboard
54 Gemini 3.1 Flash-Lite 1,336 Mar 2026 Official leaderboard
55 Gemini 2.5 Pro Jun 2025 1,285 Jun 2025 Official leaderboard

More effort adds a median 44 elo

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Muse Spark 1.3
xhigh max
1,640
Claude Opus 4.7
high
1,577
Claude Opus 4.6
high
1,543
Gemini 3.5 Flash
medium
1,504
Gemini 3 Flash
minimal
1,450
Kimi K2.5
instant thinking
1,439
GPT-5.1
high
1,421
ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 max 1,753
Claude Sonnet 5.5 xhigh 1,737
GPT-6 Astra max 1,726
Claude Fable 5.1 max 1,722
GPT-6.1 Sol max 1,702
Claude Opus 5 max 1,662
GPT-6 Sol max 1,660
Muse Spark 1.3 max 1,640
Qwen3.8 Max (0902) 1,637
Claude Fable 5 high 1,623
Qwen 3.8 Max 1,615
GPT-5.6 Sol xHigh, Codex 1,609
DeepSeek V4.1 Flash max 1,606
Grok 4.6 high 1,603
Gemini 3.7 Flash high 1,599
MiMo-V2.6 Pro 1,598
Muse Spark 1.3 xhigh 1,596
MiMo-V2.6 Flash 1,596
Gemini 3.8 Flash high 1,593
Grok 4.7 xhigh 1,586
Kimi K3 max 1,579
GLM-5.3-Flash 1,578
Claude Opus 4.7 high 1,577
Qwen 3.8 27B 1,577
GPT-6 Luna max 1,571
Claude Opus 4.7 1,562
Grok 4.5 1,562
Step 5 Preview high 1,550
Gemini 3.6 Flash high 1,545
Claude Opus 4.6 high 1,543
GPT-5.6 Terra xHigh, Codex 1,541
Claude Sonnet 5 high 1,540
Muse Spark 1.1 1,540
Claude Opus 4.6 1,538
GPT-5.5 xHigh, Codex 1,532
Claude Sonnet 4.6 1,529
Kimi K2.6 1,521
GPT-5.5 High, Codex 1,519
GPT-5.6 Luna xHigh, Codex 1,517
Kimi K2.7 Code 1,513
Seed2.1 Pro 1,510
Gemini 3.5 Flash medium 1,504
GPT-5.5 Codex 1,503
Gemini 3.5 Flash high 1,501
Gemini 3.1 Pro 1,481
Qwen 3.6 Plus 1,472
MiniMax-M3 1,470
Gemini 3 Flash 1,450
Gemini 3.5 Flash-Lite 1,448
GPT-5.3 Codex Codex 1,443
Gemini 3 Pro 1,441
Kimi K2.5 thinking 1,439
GPT-5.4 1,436
Gemini 3 Flash minimal 1,431
GLM-5V-Turbo 1,425
GPT-5.1 high 1,421
Kimi K2.5 instant 1,411
Inkling-Small 1,410
Nemotron 3.5 Super VL nvfp4 1,407
Grok 4.3 1,375
Mistral Large 3 1,361
GPT-5.1 1,351
Gemini 3.1 Flash-Lite 1,336
Gemini 2.5 Pro Jun 2025 1,285

7 models have held number one since its release in Apr 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic OpenAI
Own leaderboard 7 leaders
Claude Opus 5.5 1,753

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
1,200 1,400 1,600 1,800 2025 2026 1,200 1,400 1,600 1,800 2025 2026

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in Coding

All 243 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Opus 5.5 93.9%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471