Vals Vibe Code Bench 1-100 (Vibe Code Bench 1-100)
We show this table for reference; we do not rank on it.
Can a model extend one working web application across a long sequence of dependent requests?
Evaluation results published by Vals AI. Published leaderboard aggregates and evaluation settings only. Dataset contents are not included.
Vibe Code 1-100 score on Vibe Code Bench 1-100 — October 29, 2026
We mirror the published vibe code 1-100 score view for Vibe Code Bench 1-100. Muse Wonder Full leads the public snapshot at 33.98%, followed by Claude Opus 5.5 (30.36%) and Claude Opus 5 (28.53%). We do not use these results to rank models overall.
Muse Wonder Full
Meta
OpenHands · max reasoning
OpenHands
Claude Opus 5.5
Anthropic
OpenHands
OpenHands
Claude Opus 5
Anthropic
OpenHands
OpenHands
26 modelsCodingCurrentDisplay onlyUpdated October 29, 2026
Vibe Code 1-100 score table (26 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Muse Wonder FullMetaOpenHands · max reasoningOpenHands | 33.98% | Not reported | Closed |
| 2 | Claude Opus 5.5AnthropicOpenHandsOpenHands | 30.36% | Not reported | Closed |
| 3 | Claude Opus 5AnthropicOpenHandsOpenHands | 28.53% | Not reported | Closed |
| 4 | Claude Fable 5.1AnthropicOpenHandsOpenHands | 28.00% | Not reported | Closed |
| 5 | GPT-6 AstraOpenAIOpenHands · max reasoningOpenHands | 27.64% | Not reported | Closed |
| 6 | Muse WonderMetaOpenHands · max reasoningOpenHands | 26.55% | Not reported | Closed |
| 7 | Qwen3.8 Max InternalAlibabaOpenHandsOpenHands | 24.83% | Not reported | Unknown |
| 8 | GPT-5.6 LunaOpenAIOpenHands · max reasoningOpenHands | 22.59% | Not reported | Closed |
| 9 | Gemini 4 ArgonGoogleOpenHands · high reasoningOpenHands | 21.20% | Not reported | Closed |
| 10 | Muse Spark 1.3 MaxMetaOpenHands · max reasoningOpenHands | 20.46% | Not reported | Unknown |
| 11 | Claude Haiku 5.5AnthropicOpenHandsOpenHands | 20.41% | Not reported | Closed |
| 12 | GPT-5.6 SolOpenAIOpenHands · max reasoningOpenHands | 20.01% | Not reported | Closed |
| 13 | GLM-5.3Z.AIOpenHands · max reasoningOpenHands | 19.99% | Not reported | Open |
| 14 | Gemini 3.8 FlashGoogleOpenHands · high reasoningOpenHands | 18.77% | Not reported | Closed |
| 15 | Kimi K3Moonshot AIOpenHands · max reasoningOpenHands | 18.24% | Not reported | Open |
| 16 | DeepSeek V4 Pro 0813DeepSeekOpenHands · max reasoningOpenHands | 17.53% | Not reported | Open |
| 17 | DeepSeek V4.1 Flash 1017DeepSeekOpenHands · high reasoningOpenHands | 16.38% | Not reported | Open |
| 18 | GLM-5.3-FlashZ.AIOpenHands · max reasoningOpenHands | 16.03% | Not reported | Open |
| 19 | GPT-5.6 TerraOpenAIOpenHands · max reasoningOpenHands | 14.82% | Not reported | Closed |
| 20 | Grok 4.6xAIOpenHands · high reasoningOpenHands | 14.75% | Not reported | Closed |
| 21 | Mistral Large 4MistralOpenHands · high reasoningOpenHands | 14.26% | Not reported | Open |
| 22 | Claude Sonnet 5AnthropicOpenHandsOpenHands | 13.82% | Not reported | Closed |
| 23 | Qwen3.8 MaxAlibabaOpenHandsOpenHands | 12.83% | Not reported | Open |
| 24 | MiniMax M3MiniMaxOpenHandsOpenHands | 9.22% | Not reported | Open |
| 25 | InklingThinking Machines LabOpenHands · 0.99 reasoningOpenHands | 7.33% | Not reported | Open |
| 26 | Gemini 3.1 Pro PreviewGoogleOpenHands · high reasoningOpenHands | 6.69% | Not reported | Unknown |
How Vibe Code Bench 1-100 is shown here
BenchLM mirrors the public Vals AI Vibe Code Bench 1-100 leaderboard captured from https://www.vals.ai/benchmarks/vcb-1-100 and updated by Vals on October 29, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.
Vibe Code Bench 1-100 is display only on BenchLM. Vals proprietary or Vals-hosted aggregate views are useful context, but BenchLM does not use them as weighted ranking inputs or as a replacement for benchmark-native source records.
Snapshot
The published Vibe Code Bench 1-100 snapshot places Muse Wonder Full first at 33.98%. The third row is 5.45 points behind. The broader top-10 range is 13.52 points, so the table still separates the published systems.
26 models have been evaluated on Vibe Code Bench 1-100. The benchmark falls in the Coding category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. Vibe Code Bench 1-100 is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Vibe Code Bench 1-100
Year
2026
Tasks
Long chains of dependent web-application change requests
Format
Accuracy score
Difficulty
Long-horizon software delivery
Each run starts from a working web application and applies a long chain of dependent change requests, so an early mistake compounds through the rest of the sequence. That is a different question from the single-build Vibe Code Bench v1.1 board, which is why BenchLM keeps it on its own key. Private task set, external runner, display only.
Freshness and provenance
Version
Vibe Code Bench 1-100 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does Vibe Code Bench 1-100 measure?
Can a model extend one working web application across a long sequence of dependent requests?
Which model leads the published Vibe Code Bench 1-100 snapshot?
Muse Wonder Full currently leads the published Vibe Code Bench 1-100 snapshot with 33.98% vibe code 1-100 score. BenchLM shows this benchmark for display only and does not use it in overall rankings.
How many models are evaluated on Vibe Code Bench 1-100?
The October 29, 2026 snapshot contains 26 AI models.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.