Skip to content
← All benchmark sources

Benchmark source guide

Arena Code WebDev Benchmark

Arena Code WebDev benchmark evidence used by Vibe Bench with current ELO, votes, freshness, matching, and limits.

Stored observations
42
Matched models
42
Latest observation
Jul 16, 2026

How it is used

Method inside Vibe Bench

Vibe Bench matches a public Arena model name conservatively, normalizes its ELO to a fixed scale, and blends it into Coding only while the result is current.

Web development preference ELOSource rankVotesUncertainty

Read with care

Known limits

Arena results reflect human preference for web development outputs. Preliminary entries can move as more votes arrive.

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

ModelMeasureSource valueStatusObserved
Claude Fable 5Coding1,631VerifiedJul 16, 2026
GPT-5.6 SolCoding1,618VerifiedJul 16, 2026
GLM-5.2Coding1,587VerifiedJul 16, 2026
Claude Opus 4.8Coding1,562VerifiedJul 16, 2026
Grok 4.5Coding1,558VerifiedJul 16, 2026
Claude Opus 4.7Coding1,558VerifiedJul 16, 2026
Claude Sonnet 5Coding1,542VerifiedJul 16, 2026
Muse Spark 1.1Coding1,538PreliminaryJul 16, 2026
GLM-5.1Coding1,526VerifiedJul 16, 2026
Claude Sonnet 4.6Coding1,522VerifiedJul 16, 2026
Qwen3.7 MaxCoding1,516PreliminaryJul 16, 2026
Kimi K2.6Coding1,515VerifiedJul 16, 2026
GPT-5.5Coding1,504VerifiedJul 16, 2026
MiniMax-M3Coding1,493VerifiedJul 16, 2026
Gemini 3.5 FlashCoding1,493VerifiedJul 16, 2026
MiMo-V2.5-ProCoding1,474VerifiedJul 16, 2026
Kimi K2.7 CodeCoding1,470VerifiedJul 16, 2026
Qwen3.6 PlusCoding1,459VerifiedJul 16, 2026
DeepSeek V4 ProCoding1,459VerifiedJul 16, 2026
GPT-5.4Coding1,457VerifiedJul 16, 2026
Gemini 3.1 Pro PreviewCoding1,444VerifiedJul 16, 2026
GLM-4.7Coding1,440VerifiedJul 16, 2026
Kimi K2.5Coding1,433VerifiedJul 16, 2026
MiMo-V2.5Coding1,429VerifiedJul 16, 2026
GPT-5.4 miniCoding1,398VerifiedJul 16, 2026
Qwen3.5 397B A17BCoding1,396VerifiedJul 16, 2026
MiniMax-M2.7Coding1,395VerifiedJul 16, 2026
GPT-5Coding1,395VerifiedJul 16, 2026
GPT-5.1Coding1,391VerifiedJul 16, 2026
DeepSeek V3.2Coding1,368VerifiedJul 16, 2026
Gemma 4 31BCoding1,367VerifiedJul 16, 2026
Qwen3.5 122B A10BCoding1,364VerifiedJul 16, 2026
Gemma 4 26B A4BCoding1,361VerifiedJul 16, 2026
Grok 4.3Coding1,360VerifiedJul 16, 2026
GLM-4.6Coding1,355VerifiedJul 16, 2026
MiMo-V2-FlashCoding1,300VerifiedJul 16, 2026
Mistral Medium 3.5Coding1,267VerifiedJul 16, 2026
Mistral Large 3Coding1,224VerifiedJul 16, 2026
Gemini 2.5 ProCoding1,204VerifiedJul 16, 2026
Devstral 2Coding1,200VerifiedJul 16, 2026
Granite 4.1 8BCoding1,200VerifiedJul 16, 2026
Kimi K3Coding1,679PreliminaryJul 16, 2026