GPT-4o mini
GPT-4o mini replaces GPT-3.5 Turbo at $0.15 / $0.60 per million tokens with a 128K context and vision.
Cheapest capable OpenAI model of its time, 60% cheaper than GPT-3.5 Turbo and ahead of Claude 3 Haiku and Gemini 1.5 Flash on OpenAI-reported benchmarks. Applies the instruction-hierarchy method to resist jailbreaks and prompt injection.
- Date
- Thursday, 18 July 2024
- Lab
- OpenAI
- Kind
- model
- Access
- closed API
- Price
- $0.15/$0.60 per 1M tokens
Figures
| Measure | Value | Measured by |
|---|---|---|
| API price | $0.15 input / $0.60 output per 1M tokens 60% below GPT-3.5 Turbo; 128K input, 16K output | company |
Sources
This record was checked against its sources on 6 October 2026. How we check