Benchmark Almanac

DesignArena Frontend Web App

GPT-6 Astra 1,340 leads since September 2026
GPT-6 Astra 1,340 leads since September 2026
In use

Coding agents build multi-file React frontends from real user prompts. Users compare the running apps and vote, and the run logs record agent behavior.

One of 40 DesignArena boards. DesignArena ranks entries with Bradley-Terry ratings from user votes.

Kind
Web and app front ends
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Strongest model tested
GPT-6 Astra #3 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
Scores refreshed
November 12th, 2026
54
Entries on its own leaderboard, 53 models we track
1
Launch post lists it, from 1 lab

GPT-6 Astra leads with 1,340

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

GPT-6 Astra leads with 1,340, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.

54 entries. Scroll sideways for the rest. Bars start at 900.

  1. 1,340 GPT-6 Astra
  2. 1,318 Claude Opus 5.5
  3. 1,315 Claude Fable 5.1
  4. 1,296 Kimi K3.1 Pro
  5. 1,286 Inkling 2 Small
  6. 1,283 Qwen 3.8 Max
  7. 1,282 Muse Spark 1.3
  8. 1,279 Kimi K3
  9. 1,277 Claude Sonnet 5.5
  10. 1,273 MiMo-V2.6 Pro
  11. 1,270 GLM-5.3
  12. 1,263 GPT-6.1 Sol
  13. 1,253 Claude Opus 5
  14. 1,245 Gemini 4 Argon
  15. 1,243 Claude Fable 5
  16. 1,231 Ling 3.2 Flash
  17. 1,230 Grok 4.6
  18. 1,226 GPT-5.6 Sol
  19. 1,222 Claude Sonnet 5
  20. 1,222 Gemini 3.8 Flash
  21. 1,216 Claude Opus 4.8
  22. 1,213 GPT-6 Sol
  23. 1,210 Gemini 3.7 Flash
  24. 1,204 GLM-5.2
  25. 1,204 Muse Spark 1.2
  26. 1,199 Grok 4.5
  27. 1,192 MiniMax-M3
  28. 1,189 Qwen3.7-Max
  29. 1,188 Muse Spark 1.1
  30. 1,187 Claude Opus 4.6
  31. 1,182 Gemini 3.5 Flash
  32. 1,180 Gemini 3.6 Flash
  33. 1,177 Inkling
  34. 1,172 Claude Sonnet 4.6
  35. 1,162 GPT-6 Luna
  36. 1,158 GLM-5.1
  37. 1,158 Kimi K2.7 Code
  38. 1,141 Claude Opus 4.5
  39. 1,126 Grok 4.20
  40. 1,117 Kimi K2.5
  41. 1,108 Gemini 3.1 Pro
  42. 1,108 Grok 4.3
  43. 1,106 GLM-5V-Turbo
  44. 1,101 GPT-5.6 Terra
  45. 1,099 GPT-5.5
  46. 1,096 Gemini 3 Pro
  47. 1,092 Gemini 3 Flash
  48. 1,087 Solar Pro 4
  49. 1,067 GPT-5.2
  50. 1,041 Claude Sonnet 4.5
  51. 1,040 GPT-5.4
  52. 1,032 GPT-5.2 Codex
  53. 1,014 GPT-5.3 Codex
  54. 1,000 GPT-5.1-Codex

Bars start at 900.

  1. 1 GPT-6 Astra xhigh, max 1,340
  2. 2 Claude Opus 5.5 1,318
  3. 3 Claude Fable 5.1 1,315
  4. 4 Kimi K3.1 Pro 1,296
  5. 5 Inkling 2 Small 1,286
  6. 6 Qwen 3.8 Max 1,283
  7. 7 Muse Spark 1.3 xhigh, max 1,282
  8. 8 Kimi K3 1,279
  9. 9 Claude Sonnet 5.5 1,277
  10. 10 MiMo-V2.6 Pro 1,273
  11. 11 GLM-5.3 1,270
  12. 12 GPT-6.1 Sol 1,263
  13. 13 Claude Opus 5 1,253
  14. 14 Gemini 4 Argon 1,245
  15. 15 Claude Fable 5 1,243
  16. 16 Ling 3.2 Flash 1,231
  17. 17 Grok 4.6 1,230
  18. 18 GPT-5.6 Sol medium, xhigh 1,226
  19. 19 Claude Sonnet 5 1,222
  20. 20 Gemini 3.8 Flash 1,222
  21. 21 Claude Opus 4.8 1,216
  22. 22 GPT-6 Sol 1,213
  23. 23 Gemini 3.7 Flash 1,210
  24. 24 GLM-5.2 1,204
  25. 25 Muse Spark 1.2 1,204
  26. 26 Grok 4.5 1,199
  27. 27 MiniMax-M3 1,192
  28. 28 Qwen3.7-Max 1,189
  29. 29 Muse Spark 1.1 1,188
  30. 30 Claude Opus 4.6 1,187
  31. 31 Gemini 3.5 Flash 1,182
  32. 32 Gemini 3.6 Flash 1,180
  33. 33 Inkling 1,177
  34. 34 Claude Sonnet 4.6 1,172
  35. 35 GPT-6 Luna 1,162
  36. 36 GLM-5.1 1,158
  37. 37 Kimi K2.7 Code 1,158
  38. 38 Claude Opus 4.5 1,141
  39. 39 Grok 4.20 reasoning 1,126
  40. 40 Kimi K2.5 thinking 1,117
  41. 41 Gemini 3.1 Pro 1,108
  42. 42 Grok 4.3 1,108
  43. 43 GLM-5V-Turbo 1,106
  44. 44 GPT-5.6 Terra 1,101
  45. 45 GPT-5.5 1,099
  46. 46 Gemini 3 Pro 1,096
  47. 47 Gemini 3 Flash 1,092
  48. 48 Solar Pro 4 1,087
  49. 49 GPT-5.2 1,067
  50. 50 Claude Sonnet 4.5 1,041
  51. 51 GPT-5.4 1,040
  52. 52 GPT-5.2 Codex 1,032
  53. 53 GPT-5.3 Codex 1,014
  54. 54 GPT-5.1-Codex 1,000

Source: Official DesignArena Frontend Web App leaderboard

53 models. Scroll sideways for the rest. Bars start at 900.

  1. 1,340 GPT-6 Astra
  2. 1,318 Claude Opus 5.5
  3. 1,315 Claude Fable 5.1
  4. 1,296 Kimi K3.1 Pro
  5. 1,296 Inkling 2 Small
  6. 1,283 Qwen 3.8 Max
  7. 1,282 Muse Spark 1.3
  8. 1,279 Kimi K3
  9. 1,277 Claude Sonnet 5.5
  10. 1,273 MiMo-V2.6 Pro
  11. 1,270 GLM-5.3
  12. 1,263 GPT-6.1 Sol
  13. 1,253 Claude Opus 5
  14. 1,245 Gemini 4 Argon
  15. 1,243 Claude Fable 5
  16. 1,230 Grok 4.6
  17. 1,226 GPT-5.6 Sol
  18. 1,222 Claude Sonnet 5
  19. 1,222 Gemini 3.8 Flash
  20. 1,216 Claude Opus 4.8
  21. 1,213 GPT-6 Sol
  22. 1,210 Gemini 3.7 Flash
  23. 1,204 GLM-5.2
  24. 1,204 Muse Spark 1.2
  25. 1,199 Grok 4.5
  26. 1,192 MiniMax-M3
  27. 1,189 Qwen3.7-Max
  28. 1,188 Muse Spark 1.1
  29. 1,187 Claude Opus 4.6
  30. 1,182 Gemini 3.5 Flash
  31. 1,180 Gemini 3.6 Flash
  32. 1,177 Inkling
  33. 1,172 Claude Sonnet 4.6
  34. 1,162 GPT-6 Luna
  35. 1,158 GLM-5.1
  36. 1,158 Kimi K2.7 Code
  37. 1,141 Claude Opus 4.5
  38. 1,126 Grok 4.20
  39. 1,117 Kimi K2.5
  40. 1,108 Gemini 3.1 Pro
  41. 1,108 Grok 4.3
  42. 1,106 GLM-5V-Turbo
  43. 1,101 GPT-5.6 Terra
  44. 1,099 GPT-5.5
  45. 1,096 Gemini 3 Pro
  46. 1,092 Gemini 3 Flash
  47. 1,087 Solar Pro 4
  48. 1,067 GPT-5.2
  49. 1,041 Claude Sonnet 4.5
  50. 1,040 GPT-5.4
  51. 1,032 GPT-5.2 Codex
  52. 1,014 GPT-5.3 Codex
  53. 1,000 GPT-5.1-Codex

Bars start at 900.

  1. 1 GPT-6 Astra xhigh, max 1,340
  2. 2 Claude Opus 5.5 1,318
  3. 3 Claude Fable 5.1 1,315
  4. 4 Kimi K3.1 Pro 1,296
  5. 5 Inkling 2 Small 1,296
  6. 6 Qwen 3.8 Max 1,283
  7. 7 Muse Spark 1.3 xhigh, max 1,282
  8. 8 Kimi K3 1,279
  9. 9 Claude Sonnet 5.5 1,277
  10. 10 MiMo-V2.6 Pro 1,273
  11. 11 GLM-5.3 1,270
  12. 12 GPT-6.1 Sol 1,263
  13. 13 Claude Opus 5 1,253
  14. 14 Gemini 4 Argon 1,245
  15. 15 Claude Fable 5 1,243
  16. 16 Grok 4.6 1,230
  17. 17 GPT-5.6 Sol medium, xhigh 1,226
  18. 18 Claude Sonnet 5 1,222
  19. 19 Gemini 3.8 Flash 1,222
  20. 20 Claude Opus 4.8 1,216
  21. 21 GPT-6 Sol 1,213
  22. 22 Gemini 3.7 Flash 1,210
  23. 23 GLM-5.2 1,204
  24. 24 Muse Spark 1.2 1,204
  25. 25 Grok 4.5 1,199
  26. 26 MiniMax-M3 1,192
  27. 27 Qwen3.7-Max 1,189
  28. 28 Muse Spark 1.1 1,188
  29. 29 Claude Opus 4.6 1,187
  30. 30 Gemini 3.5 Flash 1,182
  31. 31 Gemini 3.6 Flash 1,180
  32. 32 Inkling 1,177
  33. 33 Claude Sonnet 4.6 1,172
  34. 34 GPT-6 Luna 1,162
  35. 35 GLM-5.1 1,158
  36. 36 Kimi K2.7 Code 1,158
  37. 37 Claude Opus 4.5 1,141
  38. 38 Grok 4.20 reasoning 1,126
  39. 39 Kimi K2.5 thinking 1,117
  40. 40 Gemini 3.1 Pro 1,108
  41. 41 Grok 4.3 1,108
  42. 42 GLM-5V-Turbo 1,106
  43. 43 GPT-5.6 Terra 1,101
  44. 44 GPT-5.5 1,099
  45. 45 Gemini 3 Pro 1,096
  46. 46 Gemini 3 Flash 1,092
  47. 47 Solar Pro 4 1,087
  48. 48 GPT-5.2 1,067
  49. 49 Claude Sonnet 4.5 1,041
  50. 50 GPT-5.4 1,040
  51. 51 GPT-5.2 Codex 1,032
  52. 52 GPT-5.3 Codex 1,014
  53. 53 GPT-5.1-Codex 1,000

Sources: Official DesignArena Frontend Web App leaderboard and labs’ model cards

#EntryScoreSettingDateSource
1 GPT-6 Astra 1,340 xhigh, max Sep 2026 Official leaderboard
2 Claude Opus 5.5 1,318 Sep 2026 Official leaderboard
3 Claude Fable 5.1 1,315 Sep 2026 Official leaderboard
4 Kimi K3.1 Pro 1,296 Oct 2026 Official leaderboard
5 Inkling 2 Small 1,286 Nov 2026 Official leaderboard
6 Qwen 3.8 Max 1,283 Aug 2026 Official leaderboard
7 Muse Spark 1.3 1,282 xhigh, max Sep 2026 Official leaderboard
8 Kimi K3 1,279 Jul 2026 Official leaderboard
9 Claude Sonnet 5.5 1,277 Sep 2026 Official leaderboard
10 MiMo-V2.6 Pro 1,273 Sep 2026 Official leaderboard
11 GLM-5.3 1,270 Aug 2026 Official leaderboard
12 GPT-6.1 Sol 1,263 Sep 2026 Official leaderboard
13 Claude Opus 5 1,253 Jul 2026 Official leaderboard
14 Gemini 4 Argon 1,245 Sep 2026 Official leaderboard
15 Claude Fable 5 1,243 Jun 2026 Official leaderboard
16 Ling 3.2 Flash 1,231 Official leaderboard
17 Grok 4.6 1,230 Aug 2026 Official leaderboard
18 GPT-5.6 Sol 1,226 medium, xhigh Jul 2026 Official leaderboard
19 Claude Sonnet 5 1,222 Jun 2026 Official leaderboard
20 Gemini 3.8 Flash 1,222 Sep 2026 Official leaderboard
21 Claude Opus 4.8 1,216 May 2026 Official leaderboard
22 GPT-6 Sol 1,213 Sep 2026 Official leaderboard
23 Gemini 3.7 Flash 1,210 Aug 2026 Official leaderboard
24 GLM-5.2 1,204 Jun 2026 Official leaderboard
25 Muse Spark 1.2 1,204 Aug 2026 Official leaderboard
26 Grok 4.5 1,199 Jul 2026 Official leaderboard
27 MiniMax-M3 1,192 Jun 2026 Official leaderboard
28 Qwen3.7-Max 1,189 May 2026 Official leaderboard
29 Muse Spark 1.1 1,188 Jul 2026 Official leaderboard
30 Claude Opus 4.6 1,187 Feb 2026 Official leaderboard
31 Gemini 3.5 Flash 1,182 May 2026 Official leaderboard
32 Gemini 3.6 Flash 1,180 Jul 2026 Official leaderboard
33 Inkling 1,177 Jul 2026 Official leaderboard
34 Claude Sonnet 4.6 1,172 Feb 2026 Official leaderboard
35 GPT-6 Luna 1,162 Sep 2026 Official leaderboard
36 GLM-5.1 1,158 Apr 2026 Official leaderboard
37 Kimi K2.7 Code 1,158 Jun 2026 Official leaderboard
38 Claude Opus 4.5 1,141 Nov 2025 Official leaderboard
39 Grok 4.20 1,126 reasoning Feb 2026 Official leaderboard
40 Kimi K2.5 1,117 thinking Jan 2026 Official leaderboard
41 Gemini 3.1 Pro 1,108 Feb 2026 Official leaderboard
42 Grok 4.3 1,108 Apr 2026 Official leaderboard
43 GLM-5V-Turbo 1,106 Apr 2026 Official leaderboard
44 GPT-5.6 Terra 1,101 Jul 2026 Official leaderboard
45 GPT-5.5 1,099 Apr 2026 Official leaderboard
46 Gemini 3 Pro 1,096 Nov 2025 Official leaderboard
47 Gemini 3 Flash 1,092 Dec 2025 Official leaderboard
48 Solar Pro 4 1,087 Aug 2026 Official leaderboard
49 GPT-5.2 1,067 Dec 2025 Official leaderboard
50 Claude Sonnet 4.5 1,041 Sep 2025 Official leaderboard
51 GPT-5.4 1,040 Mar 2026 Official leaderboard
52 GPT-5.2 Codex 1,032 Dec 2025 Official leaderboard
53 GPT-5.3 Codex 1,014 Feb 2026 Official leaderboard
54 GPT-5.1-Codex 1,000 Nov 2025 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-6 Astra 1,340 xhigh, max Sep 2026 Official leaderboard
2 Claude Opus 5.5 1,318 Sep 2026 Official leaderboard
3 Claude Fable 5.1 1,315 Sep 2026 Official leaderboard
4 Kimi K3.1 Pro 1,296 Oct 2026 Official leaderboard
5 Inkling 2 Small 1,296 Nov 2026 Lab model card
6 Qwen 3.8 Max 1,283 Aug 2026 Official leaderboard
7 Muse Spark 1.3 1,282 xhigh, max Sep 2026 Official leaderboard
8 Kimi K3 1,279 Jul 2026 Official leaderboard
9 Claude Sonnet 5.5 1,277 Sep 2026 Official leaderboard
10 MiMo-V2.6 Pro 1,273 Sep 2026 Official leaderboard
11 GLM-5.3 1,270 Aug 2026 Official leaderboard
12 GPT-6.1 Sol 1,263 Sep 2026 Official leaderboard
13 Claude Opus 5 1,253 Jul 2026 Official leaderboard
14 Gemini 4 Argon 1,245 Sep 2026 Official leaderboard
15 Claude Fable 5 1,243 Jun 2026 Official leaderboard
16 Grok 4.6 1,230 Aug 2026 Official leaderboard
17 GPT-5.6 Sol 1,226 medium, xhigh Jul 2026 Official leaderboard
18 Claude Sonnet 5 1,222 Jun 2026 Official leaderboard
19 Gemini 3.8 Flash 1,222 Sep 2026 Official leaderboard
20 Claude Opus 4.8 1,216 May 2026 Official leaderboard
21 GPT-6 Sol 1,213 Sep 2026 Official leaderboard
22 Gemini 3.7 Flash 1,210 Aug 2026 Official leaderboard
23 GLM-5.2 1,204 Jun 2026 Official leaderboard
24 Muse Spark 1.2 1,204 Aug 2026 Official leaderboard
25 Grok 4.5 1,199 Jul 2026 Official leaderboard
26 MiniMax-M3 1,192 Jun 2026 Official leaderboard
27 Qwen3.7-Max 1,189 May 2026 Official leaderboard
28 Muse Spark 1.1 1,188 Jul 2026 Official leaderboard
29 Claude Opus 4.6 1,187 Feb 2026 Official leaderboard
30 Gemini 3.5 Flash 1,182 May 2026 Official leaderboard
31 Gemini 3.6 Flash 1,180 Jul 2026 Official leaderboard
32 Inkling 1,177 Jul 2026 Official leaderboard
33 Claude Sonnet 4.6 1,172 Feb 2026 Official leaderboard
34 GPT-6 Luna 1,162 Sep 2026 Official leaderboard
35 GLM-5.1 1,158 Apr 2026 Official leaderboard
36 Kimi K2.7 Code 1,158 Jun 2026 Official leaderboard
37 Claude Opus 4.5 1,141 Nov 2025 Official leaderboard
38 Grok 4.20 1,126 reasoning Feb 2026 Official leaderboard
39 Kimi K2.5 1,117 thinking Jan 2026 Official leaderboard
40 Gemini 3.1 Pro 1,108 Feb 2026 Official leaderboard
41 Grok 4.3 1,108 Apr 2026 Official leaderboard
42 GLM-5V-Turbo 1,106 Apr 2026 Official leaderboard
43 GPT-5.6 Terra 1,101 Jul 2026 Official leaderboard
44 GPT-5.5 1,099 Apr 2026 Official leaderboard
45 Gemini 3 Pro 1,096 Nov 2025 Official leaderboard
46 Gemini 3 Flash 1,092 Dec 2025 Official leaderboard
47 Solar Pro 4 1,087 Aug 2026 Official leaderboard
48 GPT-5.2 1,067 Dec 2025 Official leaderboard
49 Claude Sonnet 4.5 1,041 Sep 2025 Official leaderboard
50 GPT-5.4 1,040 Mar 2026 Official leaderboard
51 GPT-5.2 Codex 1,032 Dec 2025 Official leaderboard
52 GPT-5.3 Codex 1,014 Feb 2026 Official leaderboard
53 GPT-5.1-Codex 1,000 Nov 2025 Official leaderboard

Labs report scores for 1 model in their cards. Switch the scores above to rank them together.

More effort adds a median 26 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
GPT-6 Astra
xhigh xhigh, max
1,340
Muse Spark 1.3
xhigh
1,282
GPT-5.6 Sol
medium medium, xhi…
1,226

Source: Official DesignArena Frontend Web App leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-6 Astra xhigh, max 1,340
Claude Opus 5.5 1,318
Claude Fable 5.1 1,315
GPT-6 Astra xhigh 1,314
Kimi K3.1 Pro 1,296
Inkling 2 Small 1,286
Qwen 3.8 Max 1,283
Muse Spark 1.3 xhigh, max 1,282
Kimi K3 1,279
Muse Spark 1.3 xhigh 1,279
Claude Sonnet 5.5 1,277
MiMo-V2.6 Pro 1,273
GLM-5.3 1,270
GPT-6.1 Sol 1,263
Claude Opus 5 1,253
Gemini 4 Argon 1,245
Claude Fable 5 1,243
Grok 4.6 1,230
GPT-5.6 Sol medium, xhigh 1,226
Claude Sonnet 5 1,222
Gemini 3.8 Flash 1,222
Claude Opus 4.8 1,216
GPT-6 Sol 1,213
Gemini 3.7 Flash 1,210
GLM-5.2 1,204
Muse Spark 1.2 1,204
Grok 4.5 1,199
MiniMax-M3 1,192
Qwen3.7-Max 1,189
Muse Spark 1.1 1,188
Claude Opus 4.6 1,187
Gemini 3.5 Flash 1,182
Gemini 3.6 Flash 1,180
Inkling 1,177
Claude Sonnet 4.6 1,172
GPT-5.6 Sol medium 1,172
GPT-6 Luna 1,162
GLM-5.1 1,158
Kimi K2.7 Code 1,158
Claude Opus 4.5 1,141
Grok 4.20 reasoning 1,126
Kimi K2.5 thinking 1,117
Gemini 3.1 Pro 1,108
Grok 4.3 1,108
GLM-5V-Turbo 1,106
GPT-5.6 Terra 1,101
GPT-5.5 1,099
Gemini 3 Pro 1,096
Gemini 3 Flash 1,092
Solar Pro 4 1,087
GPT-5.2 1,067
Claude Sonnet 4.5 1,041
GPT-5.4 1,040
GPT-5.2 Codex 1,032
GPT-5.3 Codex 1,014
GPT-5.1-Codex 1,000

11 models have held number one on the official leaderboard since 2023 11 models have held number one with claims included since 2023

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

Anthropic Google DeepMind Qwen Moonshot AI OpenAI
Own leaderboard 11 leaders
GPT-6 Astra 1,340

Source: Official DesignArena Frontend Web App leaderboard

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 Reported by the lab in its model card
1,000 1,100 1,200 1,300 1,400 2026 2027 1,000 1,100 1,200 1,300 1,400 2026 2027

Sources: Official DesignArena Frontend Web App leaderboard and labs’ model cards

One lab put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Thinking Machines
DateLaunch postScoreAs the post writes it
Nov 5, 2026 Inkling 2 Small 1296 DesignArena Web Apps

One lab claim next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.

Model Lab says Board says Gap
Inkling 2 Small 1,296 DesignArena Web Apps 1,286 Leaderboard +10

More in Coding

All 282 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%
  • 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%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • 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