GPT-5 Medium
| Released | 2025-08-07 |
|---|---|
| Price per 1M | $1.25 in · $10.00 out |
| Overall score | 52.9 · rank 227 |
| Context window | 400,000 tokens |
| Provider | OpenAI |
| Weights | Proprietary |
| Configuration | Medium |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 52.9 | 227 | 17 |
| Overall: Coding | 52.3 | 230 | 11 |
| Overall: Reasoning | 53.1 | 220 | 9 |
| Overall: Agentic work | 52.9 | 235 | 10 |
| Overall: Frontend | 48.7 | 234 | 5 |
| Coding | 54.2 | 192 | 2 |
| Agents | 52.4 | 182 | 3 |
| Reasoning | 52.8 | 225 | 5 |
| Math | 53.6 | 135 | 4 |
| Frontend | 44.7 | 154 | 1 |
| Multimodal | 55.7 | 67 | 2 |
| Value vs price | −0.8 | 256 |
Evidence
Coding 54.2 ±1.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 86.7% | Epoch AI | 2026-11-11 |
| SWE-bench Verified (Epoch run) | 71.5% | Epoch AI | 2026-02-05 |
Agents 52.4 ±3.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 2.0 | 49.6% | Epoch AI | 2025-11-04 |
| METR task suite | 69.6% | Epoch AI | 2026-11-11 |
| DeepResearch Bench | 48.6% | Epoch AI | 2026-11-11 |
Reasoning 52.8 ±2.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond | 85.4% | Epoch AI | 2025-08-07 |
| ARC-AGI-2 | 7.5% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 7.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 56.2% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 56.2% | Epoch AI | 2026-11-11 |
| Chess Puzzles | 32.0% | Epoch AI | 2026-10-13 |
| Mystery Game Puzzles | 15.0% | Epoch AI | 2026-08-27 |
Math 53.6 ±1.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 97.9% | Epoch AI | 2025-08-20 |
| FrontierMath Tiers 1-3 (2025 set) | 27.2% | Epoch AI | 2025-11-13 |
| FrontierMath Tier 4 (2025 set) | 6.3% | Epoch AI | 2025-08-07 |
| OTIS Mock AIME 2024-2025 | 87.2% | Epoch AI | 2025-08-07 |
Frontend 44.7 ±3.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena WebDev | 1,417 | LMArena | 2026-11-09 |
Multimodal 55.7 ±5.1
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| Terminal-Bench 2.0 | 43.8% | Epoch AI | Lower duplicate run on the same benchmark |
| Terminal-Bench 2.0 | 35.2% | Epoch AI | Lower duplicate run on the same benchmark |
| Terminal-Bench 2.0 | 33.9% | Epoch AI | Lower duplicate run on the same benchmark |
Names matched to this model
gpt-5-medium