GPT-5.1-Codex
von OpenAI · Veroeffentlicht 2025-11-13
68.6
Durchschn. Score
$1.25/1M
Eingabepreis
$10.00/1M
Ausgabepreis
400K tokens (~200 books)
Kontextfenster
multimodal
Typ
Tested on 8 benchmarks with 68.6% average. Top scores: LiveBench — Reasoning (82.0%), LiveBench — Mathematics (79.6%), LiveBench — Coding (71.8%).
Benchmark-Ergebnisse
| Benchmark | Kategorie | Score | Bar |
|---|---|---|---|
| LiveBench — Reasoning | reasoning | 82.0 | |
| LiveBench — Mathematics | math | 79.6 | |
| LiveBench — Coding | coding | 71.8 | |
| LiveBench — Language | language | 69.5 | |
| LiveBench — Overall | knowledge | 68.6 | |
| LiveBench — If | language | 63.4 | |
| LiveBench — Data Analysis | reasoning | 60.8 | |
| LiveBench — Agentic Coding | coding | 53.3 |