GPT-5 Mini (2025-08-07) — benchmark results
August 7, 2025 GPT-5 Mini snapshot, tracked when sources report the dated API model. Provider: OpenAI. Released 2025-08-07. Access: API.
Unified ELO 1645 ± 10, rank #383 of 1776 rated models, from 185 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM MedHELM - ADHD-MedEffects | 97.16 | EM | 100 |
| HELM MedHELM - CHW Care Plan | 4.96 | Jury Score | 100 |
| HELM MedHELM - MTSamples | 4.72 | Jury Score | 100 |
| HELM MedHELM - MTSamples Procedures | 4.7 | Jury Score | 100 |
| HELM MedHELM - SHC-Sequoia | 91.74 | EM | 100 |
| HELM MedHELM - STARR Patient Instructions | 4.94 | Jury Score | 100 |
| OpenVLM MMBench V1.1 CN - CP | 87.2 | Accuracy (%) | 100 |
| OpenVLM MMBench V1.1 CN - Spatial Relationship | 94.7 | Accuracy (%) | 100 |
| OpenVLM MMBench V1.1 EN - Structuralized Imagetext Understanding | 96.3 | Accuracy (%) | 100 |
| OpenVLM MMMU - Math | 86.7 | Accuracy (%) | 100 |
| OpenVLM MMMU - Finance | 93.3 | Accuracy (%) | 99.8 |
| OpenVLM MMMU - Health and Medicine | 79.3 | Accuracy (%) | 99.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-mini-2025-08-07 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.