GPT-5-Codex
| Released | 2025-09-15 |
|---|---|
| Price per 1M | $1.25 in · $10.00 out |
| Overall score | 53.6 · rank 83 |
| Context window | 400,000 tokens |
| Provider | OpenAI |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 53.6 | 83 | 3 |
| Overall: Coding | 54.5 | 75 | 3 |
| Overall: Reasoning | 53.6 | 87 | 1 |
| Overall: Agentic work | 52.3 | 98 | 3 |
| Overall: Frontend | 56.2 | 53 | 1 |
| Coding | 56.2 | 62 | 1 |
| Agents | 50.4 | 103 | 1 |
| Reasoning | 53.6 | 90 | 1 |
| Value vs price | +2.2 | 105 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 51.8 ±5.5 | 2 |
| High | 56.2 ±7.1 | 1 |
Evidence
Coding 56.2 ±5.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LiveCodeBench (Vals run) | 84.7% | Vals AI | 2026-09-01 |
Agents 50.4 ±2.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 2.0 | 44.3% | Epoch AI | 2025-09-15 |
Reasoning 53.6 ±7.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Kagi Reasoning Benchmark | 70.3% | Kagi | 2026-11-11 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| Terminal-Bench 2.0 | 43.4% | Epoch AI | A stronger configuration result was used |
| Terminal-Bench 2.0 | 41.4% | Epoch AI | A stronger configuration result was used |
Names matched to this model
gpt-5-codex