Benchmark Almanac

Vibe Code Bench 1-100

Muse Wonder Full 34% leads since October 2026
Not in model cards

Agents start from a working web app built for Vibe Code Bench v1.1. They extend it through up to ten dependent plain-English product requests without breaking it.

Later turns keep the code, data and conversation from earlier work. The board had 26 models on October 29th, 2026.

Made by
Vals AI
Released
September 2026
Kind
Building whole projects
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Strongest model tested
GPT-6 Astra #3 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
26
Models scored on its own leaderboard

Muse Wonder Full leads with 34%

Best score per model 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.

26 models. Scroll sideways for the rest.

  1. 34.0 Muse Wonder Full
  2. 30.4 Claude Opus 5.5
  3. 28.5 Claude Opus 5
  4. 28.0 Claude Fable 5.1
  5. 27.6 GPT-6 Astra
  6. 26.6 Muse Wonder
  7. 24.8 Qwen3.8 Max (0902)
  8. 22.6 GPT-5.6 Luna
  9. 21.2 Gemini 4 Argon
  10. 20.5 Muse Spark 1.3
  11. 20.4 Claude Haiku 5.5
  12. 20.0 GPT-5.6 Sol
  13. 20.0 GLM-5.3
  14. 18.8 Gemini 3.8 Flash
  15. 18.2 Kimi K3
  16. 17.5 DeepSeek V4 Pro 0813
  17. 16.4 DeepSeek V4.1 Flash
  18. 16.0 GLM-5.3-Flash
  19. 14.8 GPT-5.6 Terra
  20. 14.8 Grok 4.6
  21. 14.3 Mistral Large 4 Preview
  22. 13.8 Claude Sonnet 5
  23. 12.8 Qwen 3.8 Max
  24. 9.22 MiniMax-M3
  25. 7.33 Inkling
  26. 6.69 Gemini 3.1 Pro
  1. 1 Muse Wonder Full 34%
  2. 2 Claude Opus 5.5 30.4%
  3. 3 Claude Opus 5 28.5%
  4. 4 Claude Fable 5.1 28%
  5. 5 GPT-6 Astra 27.6%
  6. 6 Muse Wonder 26.6%
  7. 7 Qwen3.8 Max (0902) 24.8%
  8. 8 GPT-5.6 Luna 22.6%
  9. 9 Gemini 4 Argon 21.2%
  10. 10 Muse Spark 1.3 max 20.5%
  11. 11 Claude Haiku 5.5 20.4%
  12. 12 GPT-5.6 Sol 20%
  13. 13 GLM-5.3 20%
  14. 14 Gemini 3.8 Flash 18.8%
  15. 15 Kimi K3 18.2%
  16. 16 DeepSeek V4 Pro 0813 17.5%
  17. 17 DeepSeek V4.1 Flash 16.4%
  18. 18 GLM-5.3-Flash 16%
  19. 19 GPT-5.6 Terra 14.8%
  20. 20 Grok 4.6 14.8%
  21. 21 Mistral Large 4 Preview 14.3%
  22. 22 Claude Sonnet 5 13.8%
  23. 23 Qwen 3.8 Max 12.8%
  24. 24 MiniMax-M3 9.22%
  25. 25 Inkling 7.33%
  26. 26 Gemini 3.1 Pro 6.69%

Source: Official Vibe Code Bench 1-100 leaderboard

#ModelScoreSettingDateSource
1 Muse Wonder Full 34% Oct 2026 Official leaderboard
2 Claude Opus 5.5 30.4% Sep 2026 Official leaderboard
3 Claude Opus 5 28.5% Jul 2026 Official leaderboard
4 Claude Fable 5.1 28% Sep 2026 Official leaderboard
5 GPT-6 Astra 27.6% Sep 2026 Official leaderboard
6 Muse Wonder 26.6% Oct 2026 Official leaderboard
7 Qwen3.8 Max (0902) 24.8% Sep 2026 Official leaderboard
8 GPT-5.6 Luna 22.6% Jul 2026 Official leaderboard
9 Gemini 4 Argon 21.2% Sep 2026 Official leaderboard
10 Muse Spark 1.3 20.5% max Sep 2026 Official leaderboard
11 Claude Haiku 5.5 20.4% Oct 2026 Official leaderboard
12 GPT-5.6 Sol 20% Jul 2026 Official leaderboard
13 GLM-5.3 20% Aug 2026 Official leaderboard
14 Gemini 3.8 Flash 18.8% Sep 2026 Official leaderboard
15 Kimi K3 18.2% Jul 2026 Official leaderboard
16 DeepSeek V4 Pro 0813 17.5% Aug 2026 Official leaderboard
17 DeepSeek V4.1 Flash 16.4% Sep 2026 Official leaderboard
18 GLM-5.3-Flash 16% Aug 2026 Official leaderboard
19 GPT-5.6 Terra 14.8% Jul 2026 Official leaderboard
20 Grok 4.6 14.8% Aug 2026 Official leaderboard
21 Mistral Large 4 Preview 14.3% Oct 2026 Official leaderboard
22 Claude Sonnet 5 13.8% Jun 2026 Official leaderboard
23 Qwen 3.8 Max 12.8% Aug 2026 Official leaderboard
24 MiniMax-M3 9.22% Jun 2026 Official leaderboard
25 Inkling 7.33% Jul 2026 Official leaderboard
26 Gemini 3.1 Pro 6.69% Feb 2026 Official leaderboard

Accuracy against latency

The Vals scatter chart with its x axis switched from Cost to Latency: mean time per test against accuracy, on a log scale. The line joins the runs that no other run beats on both measures. We redrew this chart from the numbers on the official leaderboard.

0% 10% 20% 30% 40% 1001,00010K100K Latency (mean per test) (s), log scale, lower is better Accuracy MiniMax-M3: Accuracy 9.22%, latency (mean per test) 1,570 s Claude Sonnet 5: Accuracy 13.8%, latency (mean per test) 15K s GPT-5.6 Luna: Accuracy 22.6%, latency (mean per test) 7,375 s GPT-5.6 Sol: Accuracy 20%, latency (mean per test) 9,421 s GPT-5.6 Terra: Accuracy 14.8%, latency (mean per test) 6,536 s Inkling: Accuracy 7.33%, latency (mean per test) 3,002 s Kimi K3: Accuracy 18.2%, latency (mean per test) 6,487 s Qwen 3.8 Max: Accuracy 12.8%, latency (mean per test) 5,736 s Grok 4.6: Accuracy 14.8%, latency (mean per test) 7,851 s DeepSeek V4 Pro 0813: Accuracy 17.5%, latency (mean per test) 2,262 s GLM-5.3: Accuracy 20%, latency (mean per test) 5,170 s GLM-5.3-Flash: Accuracy 16%, latency (mean per test) 2,343 s Claude Fable 5.1: Accuracy 28%, latency (mean per test) 10K s Qwen3.8 Max (0902): Accuracy 24.8%, latency (mean per test) 9,477 s Gemini 3.8 Flash: Accuracy 18.8%, latency (mean per test) 3,931 s GPT-6 Astra: Accuracy 27.6%, latency (mean per test) 12K s DeepSeek V4.1 Flash: Accuracy 16.4%, latency (mean per test) 1,701 s Claude Opus 5.5: Accuracy 30.4%, latency (mean per test) 8,909 s Gemini 4 Argon: Accuracy 21.2%, latency (mean per test) 6,180 s Mistral Large 4 Preview: Accuracy 14.3%, latency (mean per test) 3,593 s Claude Haiku 5.5: Accuracy 20.4%, latency (mean per test) 15K s Gemini 3.1 Pro: Accuracy 6.69%, latency (mean per test) 688 s Claude Opus 5: Accuracy 28.5%, latency (mean per test) 4,905 s Muse Spark 1.3 (max): Accuracy 20.5%, latency (mean per test) 1,202 s Muse Wonder: Accuracy 26.6%, latency (mean per test) 1,868 s Muse Wonder Full: Accuracy 34%, latency (mean per test) 7,008 s Muse Wonder FullClaude Opus 5Muse WonderMuse Spark 1.3 (max)Gemini 3.1 Pro 0% 10% 20% 30% 40% 1001,00010K100K Latency (mean per test) (s), log scale, lower is better Accuracy MiniMax-M3: Accuracy 9.22%, latency (mean per test) 1,570 s Claude Sonnet 5: Accuracy 13.8%, latency (mean per test) 15K s GPT-5.6 Luna: Accuracy 22.6%, latency (mean per test) 7,375 s GPT-5.6 Sol: Accuracy 20%, latency (mean per test) 9,421 s GPT-5.6 Terra: Accuracy 14.8%, latency (mean per test) 6,536 s Inkling: Accuracy 7.33%, latency (mean per test) 3,002 s Kimi K3: Accuracy 18.2%, latency (mean per test) 6,487 s Qwen 3.8 Max: Accuracy 12.8%, latency (mean per test) 5,736 s Grok 4.6: Accuracy 14.8%, latency (mean per test) 7,851 s DeepSeek V4 Pro 0813: Accuracy 17.5%, latency (mean per test) 2,262 s GLM-5.3: Accuracy 20%, latency (mean per test) 5,170 s GLM-5.3-Flash: Accuracy 16%, latency (mean per test) 2,343 s Claude Fable 5.1: Accuracy 28%, latency (mean per test) 10K s Qwen3.8 Max (0902): Accuracy 24.8%, latency (mean per test) 9,477 s Gemini 3.8 Flash: Accuracy 18.8%, latency (mean per test) 3,931 s GPT-6 Astra: Accuracy 27.6%, latency (mean per test) 12K s DeepSeek V4.1 Flash: Accuracy 16.4%, latency (mean per test) 1,701 s Claude Opus 5.5: Accuracy 30.4%, latency (mean per test) 8,909 s Gemini 4 Argon: Accuracy 21.2%, latency (mean per test) 6,180 s Mistral Large 4 Preview: Accuracy 14.3%, latency (mean per test) 3,593 s Claude Haiku 5.5: Accuracy 20.4%, latency (mean per test) 15K s Gemini 3.1 Pro: Accuracy 6.69%, latency (mean per test) 688 s Claude Opus 5: Accuracy 28.5%, latency (mean per test) 4,905 s Muse Spark 1.3 (max): Accuracy 20.5%, latency (mean per test) 1,202 s Muse Wonder: Accuracy 26.6%, latency (mean per test) 1,868 s Muse Wonder Full: Accuracy 34%, latency (mean per test) 7,008 s Muse Wonder FullClaude Opus 5Muse WonderMuse Spark 1.3 (max)Gemini 3.1 Pro
ModelSettingAccuracyLatency (mean per test)
Muse Wonder Full 34% 7,008 s
Claude Opus 5.5 30.4% 8,909 s
Claude Opus 5 28.5% 4,905 s
Claude Fable 5.1 28% 10,359 s
GPT-6 Astra 27.6% 12,340 s
Muse Wonder 26.6% 1,868 s
Qwen3.8 Max (0902) 24.8% 9,477 s
GPT-5.6 Luna 22.6% 7,375 s
Gemini 4 Argon 21.2% 6,180 s
Muse Spark 1.3 max 20.5% 1,202 s
Claude Haiku 5.5 20.4% 14,744 s
GPT-5.6 Sol 20% 9,421 s
GLM-5.3 20% 5,170 s
Gemini 3.8 Flash 18.8% 3,931 s
Kimi K3 18.2% 6,487 s
DeepSeek V4 Pro 0813 17.5% 2,262 s
DeepSeek V4.1 Flash 16.4% 1,701 s
GLM-5.3-Flash 16% 2,343 s
GPT-5.6 Terra 14.8% 6,536 s
Grok 4.6 14.8% 7,851 s
Mistral Large 4 Preview 14.3% 3,593 s
Claude Sonnet 5 13.8% 15,370 s
Qwen 3.8 Max 12.8% 5,736 s
MiniMax-M3 9.2% 1,570 s
Inkling 7.3% 3,002 s
Gemini 3.1 Pro 6.7% 688.06 s

The best score, 34%, costs $20 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

OpenAI Qwen DeepSeek Meta Cost frontier One model at each effort level
0% 10% 20% 30% 40% $0.10 $1.00 $10 $100 $1,000 Cost per task, log scale Muse Wonder Full Muse Wonder Qwen3.8 Max (0902) GPT-5.6 Luna DeepSeek V4.1 Flash 0% 10% 20% 30% 40% $0.10 $1.00 $10 $100 $1,000 Cost per task, log scale Muse Wonder Full Muse Wonder

Source: Official Vibe Code Bench 1-100 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Muse Wonder Full 34% $20 300,595
Claude Opus 5.5 30.4% $44 804,811
Claude Opus 5 28.5% $41 395,606
Claude Fable 5.1 28% $150 1,190,214
GPT-6 Astra 27.6% $85 582,206
Muse Wonder 26.6% $14 209,675
Qwen3.8 Max (0902) 24.8% $12 381,930
GPT-5.6 Luna 22.6% $4.91 904,474
Gemini 4 Argon 21.2% $19 201,408
Muse Spark 1.3 max 20.5% $5.60 239,956
Claude Haiku 5.5 20.4% $8.30 3,593,023
GPT-5.6 Sol 20% $47 532,058
GLM-5.3 20% $8.24 325,505
Gemini 3.8 Flash 18.8% $19 372,599
Kimi K3 18.2% $9.40 156,715
DeepSeek V4 Pro 0813 17.5% $5.89 460,035
DeepSeek V4.1 Flash 16.4% $0.69 232,458
GLM-5.3-Flash 16% $0.93 267,891
GPT-5.6 Terra 14.8% $33 978,367
Grok 4.6 14.8% $26 717,171
Mistral Large 4 Preview 14.3% $3.44 263,731
Claude Sonnet 5 13.8% $71 1,985,522
Qwen 3.8 Max 12.8% $9.68 324,428
MiniMax-M3 9.22% $3.03 149,236
Inkling 7.33% $5.00 205,662
Gemini 3.1 Pro 6.69% $3.36 53,324

3 models have held number one on the official leaderboard since its release in September 2026

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

Anthropic Meta
Own leaderboard 3 leaders
Muse Wonder Full 34%

Source: Official Vibe Code Bench 1-100 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 Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Muse Wonder Full Claude Opus 5 GPT-5.6 Luna Claude Sonnet 5 MiniMax-M3 Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Muse Wonder Full GPT-5.6 Luna

Source: Official Vibe Code Bench 1-100 leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

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