GPT-5.4 (High): benchmark results
GPT-5.4 evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2026-03-05. Access: API.
Aggregate Rating 1787 ± 6, rank #165 of 2168 rated models, from 410 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BankerToolBench | 58.1 | Mean rubric score (%) | 100 |
| BankerToolBench - Deliverable - CSV | 42.8 | Mean rubric score (%) | 100 |
| BankerToolBench - Deliverable - Excel | 57.2 | Mean rubric score (%) | 100 |
| BankerToolBench - Deliverable - PDF | 56.7 | Mean rubric score (%) | 100 |
| BankerToolBench - Deliverable - PowerPoint | 57 | Mean rubric score (%) | 100 |
| BankerToolBench - Input - CSV | 58.3 | Mean rubric score (%) | 100 |
| BankerToolBench - Input - Image | 70.3 | Mean rubric score (%) | 100 |
| BankerToolBench - Input - PDF | 62.1 | Mean rubric score (%) | 100 |
| BankerToolBench - Input - PNG | 57.3 | Mean rubric score (%) | 100 |
| BankerToolBench - Input - Word | 60.7 | Mean rubric score (%) | 100 |
| BankerToolBench - Product - DCM | 41.7 | Mean rubric score (%) | 100 |
| BankerToolBench - Product - ECM | 64.2 | Mean rubric score (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-5-4-high · How It Works · Data refreshed daily, snapshot 2026-10-20.