Benchmark source guide
Artificial Analysis Benchmark Data
Artificial Analysis benchmark data used in Vibe Bench with current models, measured fields, freshness, and limits.
Stored observations
318
Matched models
117
Latest observation
Jul 17, 2026
How it is used
Method inside Vibe Bench
Vibe Bench reads the provider scores through the documented OpenRouter benchmark feed and keeps each source value before any blend.
IntelligenceCodingAgentic
Read with care
Known limits
The three broad indices summarize many tests. They are useful anchors but do not describe every coding workflow by themselves.
Every displayed result keeps its observed date and source status so preliminary or stale evidence remains visible.
Current stored evidence
Models covered by this source
| Model | Measure | Source value | Status | Observed |
|---|---|---|---|---|
| GPT-5.6 Sol | Agentic | 51.8 | Verified | Jul 17, 2026 |
| GPT-5.6 Terra | Agentic | 47.4 | Verified | Jul 17, 2026 |
| Claude Fable 5 | Agentic | 52.8 | Verified | Jul 17, 2026 |
| GPT-5.5 | Agentic | 44.9 | Verified | Jul 17, 2026 |
| Claude Opus 4.8 | Agentic | 47.2 | Verified | Jul 17, 2026 |
| Claude Opus 4.7 | Agentic | 44.4 | Verified | Jul 17, 2026 |
| Grok 4.5 | Agentic | 45.7 | Verified | Jul 17, 2026 |
| Claude Sonnet 5 | Agentic | 46.7 | Verified | Jul 17, 2026 |
| GPT-5.6 Luna | Agentic | 45.6 | Verified | Jul 17, 2026 |
| Muse Spark 1.1 | Agentic | 37.5 | Verified | Jul 17, 2026 |
| GPT-5.4 | Agentic | 41.1 | Verified | Jul 17, 2026 |
| Gemini 3.5 Flash | Agentic | 37.4 | Verified | Jul 17, 2026 |
| GLM-5.2 | Agentic | 43.1 | Verified | Jul 17, 2026 |
| Gemini 3.1 Pro Preview | Agentic | 21.4 | Verified | Jul 17, 2026 |
| Qwen3.7 Max | Agentic | 30.6 | Verified | Jul 17, 2026 |
| Claude Sonnet 4.6 | Agentic | 40.8 | Verified | Jul 17, 2026 |
| Kimi K2.6 | Agentic | 30.3 | Verified | Jul 17, 2026 |
| Kimi K2.7 Code | Agentic | 29.6 | Verified | Jul 17, 2026 |
| MiMo-V2.5-Pro | Agentic | 29.1 | Verified | Jul 17, 2026 |
| KAT Coder Pro V2 | Agentic | 15.5 | Verified | Jul 17, 2026 |
| DeepSeek V4 Pro | Agentic | 36.4 | Verified | Jul 17, 2026 |
| Nex-N2-Pro | Agentic | 31.0 | Verified | Jul 17, 2026 |
| Hy3 | Agentic | 30.7 | Verified | Jul 17, 2026 |
| MiniMax-M3 | Agentic | 35.4 | Verified | Jul 17, 2026 |
| MiMo-V2.5 | Agentic | 23.7 | Verified | Jul 17, 2026 |
| DeepSeek V4 Flash | Agentic | 31.1 | Verified | Jul 17, 2026 |
| GPT-5.4 mini | Agentic | 30.2 | Verified | Jul 17, 2026 |
| GPT-5.4 nano | Agentic | 27.5 | Verified | Jul 17, 2026 |
| Qwen3.7 Plus | Agentic | 20.8 | Verified | Jul 17, 2026 |
| GLM-5.1 | Agentic | 29.9 | Verified | Jul 17, 2026 |
| Qwen3.6 Plus | Agentic | 27.6 | Verified | Jul 17, 2026 |
| Qwen3.6 27B | Agentic | 27.0 | Verified | Jul 17, 2026 |
| MiniMax-M2.7 | Agentic | 25.6 | Verified | Jul 17, 2026 |
| Claude 4.5 Sonnet | Agentic | 24.6 | Verified | Jul 17, 2026 |
| Grok Build 0.1 0616 | Agentic | 28.0 | Verified | Jul 17, 2026 |
| MiMo-V2-Flash | Agentic | 12.0 | Verified | Jul 17, 2026 |
| GPT-5.1 | Agentic | 21.0 | Verified | Jul 17, 2026 |
| Nemotron 3 Ultra 550B A55B | Agentic | 27.4 | Verified | Jul 17, 2026 |
| Qwen3.5 397B A17B | Agentic | 19.8 | Verified | Jul 17, 2026 |
| Mistral Medium 3.5 | Agentic | 19.0 | Verified | Jul 17, 2026 |
| Kimi K2.5 | Agentic | 21.7 | Verified | Jul 17, 2026 |
| GLM-4.6 | Agentic | 17.7 | Verified | Jul 17, 2026 |
| Qwen3.5 122B A10B | Agentic | 20.7 | Verified | Jul 17, 2026 |
| GLM-4.7 | Agentic | 25.4 | Verified | Jul 17, 2026 |
| DeepSeek V3.2 | Agentic | 18.3 | Verified | Jul 17, 2026 |
| Claude 4.5 Haiku | Agentic | 16.4 | Verified | Jul 17, 2026 |
| DeepSeek V3.1 Terminus | Agentic | 18.1 | Verified | Jul 17, 2026 |
| Gemma 4 31B | Agentic | 14.4 | Verified | Jul 17, 2026 |
| Ring-2.6-1T | Agentic | 18.9 | Verified | Jul 17, 2026 |
| Grok 4.3 | Agentic | 24.1 | Verified | Jul 17, 2026 |
| Qwen3.6 35B A3B | Agentic | 21.4 | Verified | Jul 17, 2026 |
| Step 3.7 Flash | Agentic | 21.5 | Verified | Jul 17, 2026 |
| Gemma 4 26B A4B | Agentic | 11.0 | Verified | Jul 17, 2026 |
| GPT-5 | Agentic | 25.7 | Verified | Jul 17, 2026 |
| NVIDIA Nemotron 3 Super 120B A12B | Agentic | 8.7 | Verified | Jul 17, 2026 |
| Claude 4 Sonnet | Agentic | 16.6 | Verified | Jul 17, 2026 |
| North Mini Code | Agentic | 3.1 | Verified | Jul 17, 2026 |
| Qwen3 Coder Next | Agentic | 8.8 | Verified | Jul 17, 2026 |
| Gemini 3.1 Flash-Lite | Agentic | 6.2 | Verified | Jul 17, 2026 |
| Gemini 2.5 Pro | Agentic | 7.1 | Verified | Jul 17, 2026 |
| Devstral 2 | Agentic | 10.6 | Verified | Jul 17, 2026 |
| gpt-oss-120b | Agentic | 13.2 | Verified | Jul 17, 2026 |
| Devstral Small 2 | Agentic | 10.5 | Verified | Jul 17, 2026 |
| Qwen3.5 9B | Agentic | 7.4 | Verified | Jul 17, 2026 |
| Command A+ | Agentic | 9.2 | Verified | Jul 17, 2026 |
| Mistral Small 4 | Agentic | 4.7 | Verified | Jul 17, 2026 |
| Mistral Small 3.1 | Agentic | 5.2 | Verified | Jul 17, 2026 |
| Ling 2.6 Flash | Agentic | 2.3 | Verified | Jul 17, 2026 |
| DeepSeek R1 | Agentic | 3.1 | Verified | Jul 17, 2026 |
| Nova 2.0 Lite | Agentic | 3.1 | Verified | Jul 17, 2026 |
| DeepSeek V3 | Agentic | 1.6 | Verified | Jul 17, 2026 |
| Qwen3 235B A22B 2507 | Agentic | 3.8 | Verified | Jul 17, 2026 |
| DeepSeek V3 0324 | Agentic | 1.5 | Verified | Jul 17, 2026 |
| K2 Think V2 | Agentic | 1.8 | Verified | Jul 17, 2026 |
| gpt-oss-20b | Agentic | 3.1 | Verified | Jul 17, 2026 |
| Mistral Medium 3.1 | Agentic | 6.2 | Verified | Jul 17, 2026 |
| GPT-4.1 mini | Agentic | 1.7 | Verified | Jul 17, 2026 |
| Mistral Large 3 | Agentic | 5.5 | Verified | Jul 17, 2026 |
| Qwen3 Next 80B A3B | Agentic | 2.1 | Verified | Jul 17, 2026 |
| o3-mini | Agentic | 1.7 | Verified | Jul 17, 2026 |
| Llama 4 Maverick | Agentic | 1.3 | Verified | Jul 17, 2026 |
| Solar Pro 3 | Agentic | 2.7 | Verified | Jul 17, 2026 |
| Claude 3.5 Haiku | Agentic | 1.9 | Verified | Jul 17, 2026 |
| GPT-5 mini | Agentic | 19.4 | Verified | Jul 17, 2026 |
| Qwen3 32B | Agentic | 1.8 | Verified | Jul 17, 2026 |
| NVIDIA Nemotron 3 Nano 30B A3B | Agentic | 2.0 | Verified | Jul 17, 2026 |
| Ministral 3 14B | Agentic | 2.2 | Verified | Jul 17, 2026 |
| Qwen3 14B | Agentic | 1.8 | Verified | Jul 17, 2026 |
| Qwen3 30B A3B 2507 | Agentic | 1.8 | Verified | Jul 17, 2026 |
| Kimi K3 | Agentic | 50.1 | Verified | Jul 17, 2026 |
| Llama 3.3 Instruct 70B | Agentic | 0.3 | Verified | Jul 17, 2026 |
| GPT-4o mini | Agentic | 1.0 | Verified | Jul 17, 2026 |
| GPT-4.1 nano | Agentic | 1.2 | Verified | Jul 17, 2026 |
| Gemma 3 27B Instruct | Agentic | 0.3 | Verified | Jul 17, 2026 |
| Ministral 3 8B | Agentic | 1.2 | Verified | Jul 17, 2026 |
| Qwen3 8B | Agentic | 1.5 | Verified | Jul 17, 2026 |
| Llama 4 Scout | Agentic | 1.1 | Verified | Jul 17, 2026 |
| Gemma 3 12B Instruct | Agentic | 0.3 | Verified | Jul 17, 2026 |
| Llama 3.1 Instruct 8B | Agentic | 0.5 | Verified | Jul 17, 2026 |
| Ministral 3 3B | Agentic | 1.6 | Verified | Jul 17, 2026 |