Benchmark
Vals Vibe Code Bench 1-100
Coding benchmarks compare models on programming tasks. Scores are percentages.
- Models measured
- 24
- Scores recorded
- 24
- Sources
- 1
- IQ weight
- 2.17
Score distribution
- 19.2%
- Mean
- 18.8%
- Median
- 7.1%
- Std dev
- 9.2%
- 10th percentile
- 28.5%
- 90th percentile
- 6.7%
- Min
This benchmark has a weight of 2.17 for chat models, measured over 24 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Muse Wonder Full· Vals AI | 34% |
| 2 | Claude Opus 5.5· Vals AI | 30.4% |
| 3 | Claude Opus 5· Vals AI | 28.5% |
| 4 | Claude Fable 5.1· Vals AI | 28% |
| 5 | GPT-6 Astra· Vals AI | 27.6% |
| 6 | Muse Wonder· Vals AI | 26.6% |
| 7 | GPT-5.6 Luna· Vals AI | 22.6% |
| 8 | Gemini 4 Argon· Vals AI | 21.2% |
| 9 | Claude Haiku 5.5· Vals AI | 20.4% |
| 10 | GPT-5.6 Sol· Vals AI | 20% |
| 11 | GLM 5.3· Vals AI | 20% |
| 12 | Gemini 3.8 Flash· Vals AI | 18.8% |
| 13 | Kimi K3· Vals AI | 18.2% |
| 14 | DeepSeek V4 Pro 0813· Vals AI | 17.5% |
| 15 | DeepSeek V4.1 Flash· Vals AI | 16.4% |
| 16 | GLM 5.3 Flash· Vals AI | 16% |
| 17 | GPT-5.6 Terra· Vals AI | 14.8% |
| 18 | Grok 4.6· Vals AI | 14.8% |
| 19 | Mistral Large 4· Vals AI | 14.3% |
| 20 | Claude Sonnet 5· Vals AI | 13.8% |
| 21 | Qwen3.8 Max· Vals AI | 12.8% |
| 22 | MiniMax M3· Vals AI | 9.2% |
| 23 | Inkling FP4· Vals AI | 7.3% |
| 24 | Gemini 3.1 Pro Preview· Vals AI | 6.7% |