Model positioning
OpenAI o4-mini is a small reasoning model in the o-series, optimized for fast, economical reasoning, and particularly suited to math, coding, and visual tasks. (openai.com)
Capabilities and uses
- Supports text and image input and text output, and can be used for chart and image analysis, math problem-solving, code writing, and data analysis.
- Supports function calling and structured outputs; at launch, it could use tools to complete multistep tasks, such as retrieving information or running code.
- The model documentation lists a context window of 200,000 tokens and a maximum output of 100,000 tokens; the knowledge cutoff date is June 1, 2024. (developers.openai.com)
Limitations and availability
- Does not support audio or video input, and does not directly output audio or video. Model responses may be incorrect, and important results should be verified.
- OpenAI's announcement now states that the model is no longer available in ChatGPT;
o4-miniand the snapshoto4-mini-2025-04-16in the API were shut down on October 23, 2026. The replacement model recommended in the official deprecation table isgpt-5.6-terra. (developers.openai.com)
Pricing
OpenAI's official model documentation still lists text token prices for the base model: 1.10 US dollars (USD) per 1 million input tokens, 0.275 US dollars per 1 million cached input tokens, and 4.40 US dollars per 1 million output tokens. These are the listed prices for the base model and do not include potential additional charges such as tool calls; the model has been shut down, so its current availability cannot be inferred from these prices. Source: OpenAI o4-mini model and pricing documentation, accessed November 18, 2026. (developers.openai.com)
Release and official sources
- Release date: April 16, 2025.
- OpenAI official release announcement (openai.com)
- OpenAI API deprecation schedule (developers.openai.com)