GPT-5.4 High
| Released | 2026-03-05 |
|---|---|
| Price per 1M | $2.50 in · $15.00 out |
| Overall score | 57.9 · rank 150 |
| Context window | 1,050,000 tokens |
| Provider | OpenAI |
| Weights | Proprietary |
| Configuration | High |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 57.9 | 150 | 15 |
| Overall: Coding | 55.2 | 188 | 11 |
| Overall: Reasoning | 60.8 | 92 | 7 |
| Overall: Agentic work | 58.1 | 156 | 10 |
| Overall: Frontend | 51.5 | 198 | 6 |
| Coding | 56.2 | 161 | 4 |
| Reasoning | 61.1 | 107 | 6 |
| Math | 60.4 | 47 | 1 |
| Frontend | 47.7 | 130 | 1 |
| General | 58.8 | 77 | 2 |
| Multimodal | 60.0 | 35 | 1 |
| Value vs price | +2.7 | 172 |
Evidence
Coding 56.2 ±1.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MirrorCode | 15.6% | Epoch AI | 2026-08-10 |
| GSO | 25.5% | Epoch AI | 2026-11-11 |
| SWE-bench Verified (Epoch run) | 76.9% | Epoch AI | 2026-03-06 |
| ProgramBench | 0.5% | Vals AI | 2026-11-09 |
Reasoning 61.1 ±1.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Extended NYT Connections | 90.6% | Lech Mazur | 2026-11-10 |
| ARC-AGI-2 | 67.5% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 67.5% | ARC Prize | 2026-11-11 |
| GPQA Diamond | 89.9% | Epoch AI | 2026-07-15 |
| ARC-AGI-1 (ARC Prize verified) | 92.7% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 92.7% | Epoch AI | 2026-11-11 |
| ARC-AGI-3 | 0.2% | ARC Prize | 2026-11-11 |
| Chess Puzzles | 42.0% | Epoch AI | 2026-10-13 |
Math 60.4 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 97.8% | Epoch AI | 2026-07-15 |
Frontend 47.7 ±3.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena WebDev | 1,462 | LMArena | 2026-11-09 |
General 58.8 ±2.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,540 | Lech Mazur | 2026-10-26 |
| LMArena Text | 1,475 | LMArena | 2026-11-09 |
Multimodal 60.0 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,282 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| LMArena WebDev | 1,462 | LMArena | Lower duplicate run on the same benchmark |
Names matched to this model
gpt-5-4-high