BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

Vibe Code Bench leaderboard

As of 6 Nov 2026, the top score on Vibe Code Bench is 92.4% by Claude Sonnet 5.5 (Anthropic), ahead of Gemini 4 Argon at 91.9%, across 122 model configurations with a published result (Vals AI, result dated 5 Nov 2026). From a product brief the agent builds a working web app, judged on function and completeness. Vibe Code Bench is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 6 Nov 2026)

  • Claude Sonnet 5.5 holds the top Vibe Code Bench v1.1 score at 92.4% (Vals AI, 5 Nov 2026).
  • Gemini 4 Argon is second on Vibe Code Bench v1.1 at 91.9% (Vals AI, 5 Nov 2026).
  • Vibe Code Bench v1.1: Building web apps from scratch from a brief. Run by Vals AI (Vals AI).
  • 122 model configurations have a published Vibe Code Bench v1.1 result on BenchLeader (BenchLeader, 6 Nov 2026).
  • Vibe Code Bench v1.1 does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Building web apps from scratch from a brief. Run by Vals AI.

Published by
Vals AI
Category
Coding
Index weight
Reference only
Models
122
Data as of
6 Nov 2026

Vals AI (vals.ai).

What the test looks like

From a product brief the agent builds a working web app, judged on function and completeness.

How it is scored

Rubric score, run by Vals AI.

What to keep in mind

Rubric-graded; design choices and framework familiarity affect scores.

Top 15
  1. 1Claude Sonnet 5.592.4%
  2. 2Gemini 4 Argon (high)91.9%
  3. 3Claude Haiku 5.590.4%
  4. 4Claude Fable 590.3%
  5. 5Claude Opus 5.590.3%
  6. 6Claude Fable 5.190.3%
  7. 7GPT-6 Astra (max)89.6%
  8. 8Gemini 4 Carbon (high)89.1%
  9. 9GPT-6.1 Sol (max)88.9%
  10. 10Claude Opus 588.4%
  11. 11GPT-6 Sol (max)87.8%
  12. 12DeepSeek V4.1 Flash 1017 (max)87.7%
  13. 13Grok 4.7 (xhigh)86.2%
  14. 14Muse Wonder Full (max)86.0%
  15. 15Muse Spark 1.3 (max)85.9%
122 of 122
#
1Claude Sonnet 5.592.4%–$31.252026-11-05
2Gemini 4 Argonhigh91.9%66.0$8.112026-11-05
3Claude Haiku 5.590.4%–$6.072026-11-05
4Claude Fable 590.3%66.5$41.712026-11-05
5Claude Opus 5.590.3%63.4$57.922026-11-05
6Claude Fable 5.190.3%66.4$33.372026-11-05
7GPT-6 Astramax89.6%67.8$38.512026-11-05
8Gemini 4 Carbonhigh89.1%69.2$8.252026-11-05
9GPT-6.1 Solmax88.9%67.3$6.232026-11-05
10Claude Opus 588.4%65.2$33.882026-11-05
11GPT-6 Solmax87.8%63.3$26.362026-11-05
12DeepSeek V4.1 Flash 1017max87.7%59.2$0.5802026-11-05
13Grok 4.7xhigh86.2%58.3$15.832026-11-05
14Muse Wonder Fullmax86.0%62.9$2.292026-11-05
15Muse Spark 1.3max85.9%62.0$2.542026-11-05
16MiMo-V2.6-Pro85.2%57.3$1.042026-11-05
17Kimi K3max85.0%62.1$10.012026-11-05
18Kimi K3.1 Promax84.9%–$7.492026-11-05
19DeepSeek V4.1 Flashhigh84.7%55.7$0.4072026-11-05
20GPT-6.1 Terramax84.0%65.3$6.002026-11-05
21Ember-183.9%–$8.562026-11-05
22Muse Wondermax83.1%62.3$1.802026-11-05
23Muse Spark 1.3xhigh82.9%58.5$2.102026-11-05
24Claude Opus 4.882.7%60.0$26.882026-11-05
25DeepSeek V4 Promax82.3%57.0$0.3562026-11-05
26GPT-6 Lunamax81.7%56.7$1.352026-11-05
27Mistral Large 4 OSShigh81.6%–$11.722026-11-05
28NewInkling 2 Small81.5%57.4$1.902026-11-05
29Claude Sonnet 581.3%55.9$25.392026-11-05
30Hy481.0%55.7$2.392026-11-05
31GPT-5.6 Solmax80.5%65.4$33.402026-11-05
32Muse Spark 1.2xhigh79.1%61.7$1.532026-11-05
33MiMo-V2.6-Flash79.0%55.0$0.5632026-11-05
34Gemini 3.8 Flashhigh78.7%62.0$6.872026-11-05
35Mistral Large 4high78.4%49.5$20.282026-11-05
36GLM 5.3max78.1%59.0$12.452026-11-05
37Claude Opus 4.8 Claude Code77.5%–$6.862026-11-05
38Qwen3.8 Max Internal77.1%–$11.672026-11-05
39GPT-5.6 Lunamax77.1%56.3$0.7262026-11-05
40Grok 4.6high76.2%60.5$4.882026-11-05
41Muse Spark Openxhigh76.2%54.0$1.662026-11-05
42DeepSeek V4 Flashhigh74.7%57.3$0.9062026-11-05
43GPT-5.6 Terramax74.6%60.7$7.892026-11-05
44MiniMax M3.1 Flash73.9%53.0$1.152026-11-05
45Muse Spark 1.1xhigh72.2%59.5$0.9232026-11-05
46Claude Opus 4.771.0%63.1$21.412026-11-05
47Gemini 3.7 Flashhigh70.4%62.2$4.832026-11-05
48Step 5 Preview70.4%54.9$5.862026-11-05
49GPT-5.5xhigh69.8%64.3$16.662026-11-05
50Grok 4.5high69.0%56.9$4.102026-11-05
51GPT-5.4xhigh67.4%63.2$10.692026-11-05
52GPT 5.5 Factory67.4%–$5.932026-11-05
53Beam 1.066.4%45.8–2026-11-05
54Qwen3.8 27Bxhigh64.8%52.5$23.462026-11-05
55Qwen3.8 Maxmax64.7%58.8$8.242026-11-05
56Gemini 3.6 Flashhigh64.0%61.2$3.042026-11-05
57GLM 5.2max64.0%59.5$8.492026-11-05
58GPT-5.3-Codexxhigh61.8%62.9$11.912026-11-05
59Mercury 3high60.9%–$0.6272026-11-05
60GPT 5.5 Codex58.2%–$3.632026-11-05

More coding benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Vibe Code Bench v1.1

Cite as: BenchLeader, “Vibe Code Bench v1.1 leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/vals_vibe_code.

Vibe Code Bench v1.1: questions

What does Vibe Code Bench v1.1 measure?
From a product brief the agent builds a working web app, judged on function and completeness. Scores are reported in percent of tasks solved; higher is better.
What is the top score on Vibe Code Bench?
As of 6 Nov 2026, the highest score on Vibe Code Bench v1.1 is 92.4% by Claude Sonnet 5.5 (result dated 5 Nov 2026), ahead of Gemini 4 Argon at 91.9%. Scores come from Vals AI and are re-read daily.
Which AI model leads Vibe Code Bench v1.1?
Claude Sonnet 5.5 leads Vibe Code Bench v1.1 with 92.4% as of 6 Nov 2026 (result dated 5 Nov 2026), ahead of Gemini 4 Argon at 91.9%.
How many models have Vibe Code Bench v1.1 results?
122 model configurations have a Vibe Code Bench v1.1 result on BenchLeader, all taken from Vals AI.
Who runs Vibe Code Bench v1.1 and how often is it updated?
Vibe Code Bench v1.1 is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
Does Vibe Code Bench v1.1 count toward the BenchLeader Index?
No. Vibe Code Bench v1.1 is shown for reference but left out of the composite index.