Benchmark Performance
| Benchmark | Score | Unit | Evaluated At | Notes | Source |
|---|
| MMLU | 79.7 | % | 2024-07-18 | 5-shot | view |
| HUMANEVAL | 78 | pass@1 | 2024-07-18 | — | view |
| GSM8K | 75.1 | % | 2024-07-18 | 0-shot CoT | view |
| MATH | 51.8 | % | 2024-07-18 | 0-shot CoT | view |
| BBH | 70.3 | % | 2024-07-18 | 3-shot CoT | view |
| GPQA | 42.6 | % | 2024-07-18 | 0-shot | view |
| IFEVAL | 67.9 | % | 2024-07-18 | prompt_strict | view |
| ARC | 93.9 | % | 2024-07-18 | challenge | view |
| MUSR | 53.1 | % | 2024-07-18 | 0-shot | view |
| WINOGRANDE | 79.6 | % | 2024-07-18 | 0-shot | view |
Pricing
| Tier | Price | Currency |
|---|
| Input | $0.15 / Mtok | USD |
| Output | $0.6 / Mtok | USD |
| Cache Read | $0 / Mtok | USD |
| Cache Write | $0 / Mtok | USD |
Source:
https://openai.com/api/pricing/
· as of 2024-07-18
Compliance
- Data Residency: US
- SOC2: ✓
- HIPAA: ✓
- GDPR: ✓
- ISO 27001: ✓
GPT-4o mini
Modellübersicht
OpenAI 经济型多模态模型, 128K 上下文, 支持文本与图像输入, 平衡成本与性能, 适合大规模部署。
Kernspezifikationen
| Anbieter | Version | Veröffentlichungsdatum | Kontextfenster | Eingabemodalitäten | Ausgabemodalitäten | Lizenz |
|---|
| Openai | 4o-mini | 2024-07-18 | 128K | text, image | text | Proprietary |
Benchmark-Leistung
| Benchmark | Ergebnis | Einheit | Notizen |
|---|
| MMLU (Massive Multitask Language Understanding) | 79.7 | % | 5-shot |
| HumanEval | 78.0 | pass@1 | — |
| GSM8K (Grade School Math 8K) | 75.1 | % | 0-shot CoT |
| MATH | 51.8 | % | 0-shot CoT |
| BBH (BIG-Bench Hard) | 70.3 | % | 3-shot CoT |
| GPQA | 42.6 | % | 0-shot |
| IFEval | 67.9 | % | prompt_strict |
| ARC | 93.9 | % | challenge |
| MUSR | 53.1 | % | 0-shot |
| WinoGrande | 79.6 | % | 0-shot |
Preise
| Eingabe | Ausgabe | Cache-Lesen | Cache-Schreiben |
|---|
| — | — | — | — |
pro Million Token
Stärken
Schwächen
Anwendungsfälle
Referenzen