← All models

GPT-4o mini search preview

Published Publisher OpenAIgpt-4o-mini-search-preview

Model positioning

GPT-4o mini search preview is a fast, low-cost small preview model for web search, specifically trained to understand and execute web search queries in the Chat Completions API. Its fixed version is gpt-4o-mini-search-preview-2025-03-11. (developers.openai.com)

Capabilities and uses

  • Supports text input and output and can be used for question answering and information retrieval based on web information; OpenAI states that answers generated using API web search can include source links. (developers.openai.com)
  • The context window is 128,000 tokens, and the maximum output is 16,384 tokens; the knowledge cutoff date is October 1, 2023. Supports streaming and structured outputs. (developers.openai.com)

Limitations and status

Image, audio, or video input and output are not supported, nor are function calling, fine-tuning, or predicted outputs. The model has been shut down: the official page lists July 23, 2026 as the shutdown date for the fixed version and lists gpt-5.6-terra as the replacement model. (developers.openai.com)

Release date

OpenAI announced the model on March 11, 2025. (developers.openai.com)

Pricing

Date checked: November 19, 2026. The following are US dollar (USD) prices listed on the official pages; the model has been shut down, so these cannot be regarded as current purchase quotes.

  • Text tokens: $0.15 per million input tokens and $0.60 per million output tokens. Source: Official model and pricing page. (developers.openai.com)
  • Web search: In addition to token charges, search queries incur a tool-call fee. The release announcement states that search pricing for this model starts at $25 per thousand queries; this is the starting price in the announcement and does not represent the price for all search configurations. Source: OpenAI release announcement. (developers.openai.com)

Official sources