Model Positioning
GPT-4.1 nano is a model in the GPT-4.1 family that emphasizes low latency and low cost. Its fixed version ID is gpt-4.1-nano-2025-04-14. (developers.openai.com)
Capabilities and Uses
- Supports text and image input and text output, and excels at instruction following and tool calling; supports function calling, structured outputs, and streaming. (developers.openai.com)
- The context window is 1,047,576 tokens, and the maximum output is 32,768 tokens. (developers.openai.com)
- Use cases cited in the official announcement include classification and autocompletion. It can also be used for cost- or latency-sensitive applications such as long-document processing. (openai.com)
Limitations and Status
- The knowledge cutoff date is June 1, 2024; audio or video input and output are not supported. The model does not have a separate reasoning step, and complex tasks may require a more capable model. (developers.openai.com)
- OpenAI's deprecation record shows that
gpt-4.1-nanoand its 2025-04-14 fixed version were shut down on October 23, 2026; the replacement model listed in the record isgpt-5.6-luna. (developers.openai.com)
Release and Pricing
- Release date: April 14, 2025. (openai.com)
- Historical pricing: The OpenAI API standard text token prices listed in the release announcement were 0.10 US dollars (USD) per million input tokens, 0.025 US dollars per million cached input tokens, and 0.40 US dollars per million output tokens. The announcement also stated that the Batch API offered an additional 50% discount at the time. The model has now been shut down, and these amounts are for historical reference only and do not represent currently available purchase prices. Pricing source: OpenAI release announcement, access date: November 18, 2026. (openai.com)