GPT-5.2-Codex
von OpenAI · Veroeffentlicht 2026-01-14
70.6
Durchschn. Score
$1.75/1M
Eingabepreis
$14.00/1M
Ausgabepreis
400K tokens (~200 books)
Kontextfenster
multimodal
Typ
Tested on 9 benchmarks with 70.6% average. Top scores: LiveBench — Mathematics (88.8%), LiveBench — Coding (83.6%), LiveBench — Data Analysis (78.2%).
Benchmark-Ergebnisse
| Benchmark | Kategorie | Score | Bar |
|---|---|---|---|
| LiveBench — Mathematics | math | 88.8 | |
| LiveBench — Coding | coding | 83.6 | |
| LiveBench — Data Analysis | reasoning | 78.2 | |
| LiveBench — Reasoning | reasoning | 77.7 | |
| LiveBench — Overall | knowledge | 74.3 | |
| LiveBench — Language | language | 73.7 | |
| LiveBench — If | language | 66.5 | |
| LiveBench — Agentic Coding | coding | 51.7 | |
| SWE-Bench Pro (Public) | agentic | 41.0 |