TestsBenchmarks · Coding
Vals Vibe Code Bench
Share of from-scratch web applications an agent builds that pass end-to-end functional checks.
Measured by independent testersIndependent Reported by the makerVendor-reported
| Thinking levelSetting | |||||||
|---|---|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5 | Anthropic | 92.4% | Maxeffort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 2 | Gemini 4 Argon | Google (Gemini / DeepMind) | 91.9% | Highreasoning_effort=high | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 3 | Claude Fable 5 | Anthropic | 90.3% | Maxeffort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 4 | Claude Opus 5.5 | Anthropic | 90.3% | Maxeffort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 5 | Claude Fable 5.1 | Anthropic | 90.3% | Maxeffort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 6 | GPT-6 Astra | OpenAI | 89.6% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 7 | GPT-6.1 Sol | OpenAI | 88.9% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 8 | Claude Opus 5 | Anthropic | 88.4% | Default | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 9 | GPT-6 Sol | OpenAI | 87.8% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 10 | Grok 4.7 | SpaceXAI (formerly xAI) | 86.2% | Extra highreasoning_effort=xhigh | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 11 | Muse Spark 1.3 | Meta (Meta Superintelligence Labs, Muse) | 85.9% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 12 | MiMo-V2.6-Pro | Xiaomi MiMo | 85.2% | Default | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 13 | Kimi K3 | Moonshot AI (Kimi) | 85.0% | Default | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 14 | DeepSeek V4.1 Flash | DeepSeek | 84.7% | Highreasoning_effort=high | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 15 | Claude Opus 4.8 | Anthropic | 82.7% | Maxeffort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 16 | DeepSeek V4 Pro (0813) | DeepSeek | 82.3% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 17 | GPT-6 Luna | OpenAI | 81.7% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 18 | Claude Sonnet 5 | Anthropic | 81.3% | Maxeffort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 19 | GPT-5.6 Sol | OpenAI | 80.5% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 20 | Muse Spark 1.2 | Meta (Meta Superintelligence Labs, Muse) | 79.1% | Extra highreasoning_effort=xhigh | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 21 | MiMo-V2.6-Flash | Xiaomi MiMo | 79.0% | Default | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 22 | Gemini 3.8 Flash | Google (Gemini / DeepMind) | 78.7% | Highreasoning_effort=high | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |
| 23 | GLM-5.3 | Z.ai (Zhipu) | 78.1% | Maxreasoning_effort=max | Independent testIndependentVals.ai ↗ | OpenHands | 29 Sep 2026 |