Skip to main content
Evaluation sources/Vibe Code Bench 1–100
ObservingCoding

Vibe Code Bench 1–100

Vals AI · Public model evaluation results from Vibe Code Bench 1–100.

Official evaluation
Source description

This source measures concrete delivery performance in Coding; raw scores are shown separately from the normalized capability score.

Limitation: Different versions, reasoning settings, and test sets may change scores; read them with the data date.

How it is used

Participating sources affect capability scores; reference, cross-reference, and observing sources are for comparison and verification only.

Model results from this source

Claude Opus 5.530.3630.42026-11-05
Claude Opus 528.5328.52026-11-05
Claude Fable 5.12828.02026-11-05
GPT-6 Astra27.6427.62026-11-05
GPT-5.6 Luna22.5922.62026-11-05
Gemini 4 Argon21.221.22026-11-05
Muse Spark 1.320.4620.52026-11-05
Claude Haiku 5.520.4120.42026-11-05
GPT-5.6 Sol20.0120.02026-11-05
GLM 5.319.9920.02026-11-05
Gemini 3.8 Flash18.7718.82026-11-05
Kimi K318.2418.22026-11-05
DeepSeek V4 Pro 081317.5317.52026-11-05
DeepSeek V4.1 Flash16.3816.42026-11-05
GLM-5.3 Flash16.0316.02026-11-05
GPT-5.6 Terra14.8214.82026-11-05
Grok 4.614.7514.82026-11-05
Mistral Large 414.2614.32026-10-22
Claude Sonnet 513.8213.82026-11-05
Qwen3.8 Max12.8312.82026-11-05
MiniMax-M39.229.22026-11-05
Inkling7.337.32026-11-05

Frequently asked questions