AI Radar

GPT-4o mini

Available · Language, Multimodal

GPT-4o mini is a small, low-cost OpenAI model released on 18 July 2024 in the API and in ChatGPT, where it took the place of GPT-3.5. It accepts text and image input with a 128K context window, and OpenAI says it is the first model to apply its instruction hierarchy method against jailbreaks and prompt injections. [1] Primary source

Timeline of GPT-4o mini →

Claims and evidence

  • Released Primary source[1]Availability and pricing [2]Jul 18, 2024
  • Fine-tuning for GPT-4o mini launched Primary source[2]Jul 23, 2024
  • Replaced by GPT-4.1 mini in ChatGPT Primary source[3]Introducing GPT-4.1 mini, replacing GPT-4o mini, in ChatGPT for all users (May 14, 2025)
  • Status AvailableNot listed on the API deprecations page when checked on 2026-10-01. In ChatGPT, GPT-4.1 mini took its place on 2025-05-14. Primary source[4]Model details; Snapshots [5] [3]Introducing GPT-4.1 mini, replacing GPT-4o mini, in ChatGPT for all users (May 14, 2025)
  • Successor of GPT-3.5 TurboGPT-4o mini replaced GPT-3.5 in ChatGPT at launch, and the GPT-3.5 Turbo model page recommends gpt-4o-mini in its place. Primary source[1]Availability and pricing [6]description
  • Change · Modality Accepts image input as well as text in the API.Compared with GPT-3.5 Turbo Primary source[1]
  • Change · Context length Context window of 128,000 tokens with up to 16,384 output tokens, up from 16,385 tokens for gpt-3.5-turbo-0125.Compared with GPT-3.5 Turbo Primary source[4] [6]
  • Change · Efficiency Priced at 15 cents per million input tokens and 60 cents per million output tokens, which OpenAI says is more than 60 percent below GPT-3.5 Turbo.Compared with GPT-3.5 Turbo Primary source[1]
  • Change · Safety OpenAI says it is the first model to apply its instruction hierarchy method, meant to resist jailbreaks, prompt injections and system prompt extraction.Compared with GPT-3.5 Turbo Primary source[1]
  • Input text, imageAt launch the announcement said support for video and audio inputs and outputs would come later. Primary source[4]Model details [1]third paragraph
  • Output text Primary source[4]Model details
  • Feature Function calling Primary source[4]Supported features [1]A small model with superior textual intelligence and multimodal reasoning
  • Feature Fine tuning available Primary source[2]Jul 23, 2024 [4]Supported features
  • Context window 128,000 tokensThe announcement gives 128K. Primary source[4]Model details [1]third paragraph
  • Parameters Not disclosedNot disclosed by OpenAI. No claim madeNo source
  • Access API, consumer app Primary source[1]Availability and pricing

Lineage

Predecessors

Successors

No known successor.

Based on

Not derived from another model.

Variants and derived

None recorded.

Siblings

None recorded.

All ancestors

All descendants

None.

Variants

No variants recorded in this record.

Related AI Radar coverage

AI Radar coverage starts in June 2026; no coverage linked yet.

All model releases from OpenAI on AI Radar →