LiveCodeBench Pro — Elo-scale metric (2393-2887), incompatible with % LiveCodeBench variants - diff scale, never merge.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Gemini 3.1 Pro | 2887 | 1 | 2026-05-01 | |
| 2 | Gemini 3 Pro | 2439 | 1 | 2026-05-01 | |
| 3 | GPT-5.2 | OpenAI | 2393 | 1 | 2026-05-01 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.