#20
GPT-5.3 Codex
OpenAI · Proprietary · Reasoning · 400K context
64/ 100 composite
Released 2026-02-05
Benchmark Scores by Category
Coding
58.5
Agentic
64.1
Scores based on weighted benchmarks: SWE-bench, LiveCodeBench (Coding); HLE, MMLU-Pro (Knowledge); ARC-AGI-2, LongBench v2 (Reasoning); Terminal-Bench, BrowseComp (Agentic); FrontierMath, AIME (Math).
API Pricing
Input
$1.75
per 1M tokens
Output
$14
per 1M tokens
Value Score
4.7
score per $1 output
Estimate Your Costs
Use our cost calculators to estimate monthly spend based on your specific workload: