GPT-5.4 (Low): benchmark results
GPT-5.4 evaluated at the low reasoning-effort setting. Provider: OpenAI. Released 2026-03-06. Access: API.
Unified ELO 1664 ± 1, rank #202 of 1761 rated models, from 44 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Natural Intelligence | 67.6 | NatInt Score | 97 |
| UGI - Writing | 64.85 | Writing Score | 95.5 |
| AA Omniscience - Software Engineering (SWE) | 79.9 | Accuracy (%) | 94.7 |
| AA Omniscience - Health | 45.4 | Accuracy (%) | 93.3 |
| AA Terminal-Bench Hard | 43.18 | Accuracy (%) | 90.3 |
| AA-Omniscience Accuracy | 47.85 | Accuracy (%) | 90.3 |
| AA Omniscience - Law | 39.3 | Accuracy (%) | 88.5 |
| AA Omniscience - Humanities & Social Sciences | 42 | Accuracy (%) | 88 |
| AA Omniscience - Business | 35.9 | Accuracy (%) | 87.8 |
| IH-Benchmark | 94.1 | Overall Compliance (%) | 86.1 |
| AA Omniscience | 4.78 | Score | 85.8 |
| AA Omniscience - Science, Engineering & Mathematics | 44.6 | Accuracy (%) | 85.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-4-low · How It Works · Data refreshed daily, snapshot 2026-09-05.