o4-mini-deep-research

Retired Jul 2026

A cost-efficient o4-mini variant, released June 2025, fine-tuned for autonomous research with capabilities similar to o3-deep-research.

Context: 200K tokens
Max output: 100K
Cutoff: Jun 2024
Released: Jun 2025

Visit OpenAI Compare vs other LLMs →

Lium

Built for agents: CLI, SDK and MCP server. Your agent rents and runs GPUs itself.

A permissionless marketplace. 1000+ nodes from independent providers worldwide.

Hard to find nodes available; B200 and B300 Blackwell nodes listed now.

Consumer + DC grade cards for cheap. H100s from $1/hr, RTX 5090s from $0.35/hr.

View Pricing

Capabilities

What o4-mini-deep-research accepts and can do, as published by OpenAI.

Text Accepts and generates natural-language text.
Image Accepts images as input alongside text.
Reasoning Thinks before it answers, always on or as a switchable mode.
Web search Search the web for up-to-date information.
Code execution Write and execute code in a sandboxed environment.
File upload Process uploaded documents, spreadsheets, and other files.
MCP Connect to services via Model Context Protocol.

Common questions

What is o4-mini-deep-research good for?

Autonomous research tasks that search and synthesize web and private sources: it was the lighter of OpenAI's two deep research models.

When is o4-mini-deep-research not a good fit?

Anything new: OpenAI retired it in July 2026. It had no function calling or structured output, took no audio or video input, and its knowledge is dated.

Can I self-host o4-mini-deep-research?

No. OpenAI does not publish the weights; o4-mini-deep-research is available only through hosted APIs.