{
 "slug": "GPT-4o_mini",
 "title": "GPT-4o mini",
 "type": "model",
 "short_desc": "OpenAI's July 2024 small multimodal model, a low-cost successor to GPT-3.5 Turbo in ChatGPT and the API.",
 "categories": [
  "OpenAI models",
  "Models",
  "Multimodal models",
  "2024 model releases"
 ],
 "infobox": {
  "Developer": "OpenAI",
  "Family": "GPT",
  "Announced": "July 2024",
  "Architecture": "Transformer-based multimodal model; details undisclosed",
  "Parameters": "Undisclosed",
  "Context window": "128,000 tokens (up to 16,384 output tokens)",
  "Post-training": "Instruction tuning, RLHF, instruction-hierarchy training",
  "License / access": "Proprietary; API and ChatGPT",
  "Paper/report": "OpenAI announcement post (July 2024); covered by the GPT-4o system card"
 },
 "infobox_links": {
  "Developer": [
   "OpenAI"
  ],
  "Family": [
   "GPT_(model_family)"
  ],
  "Architecture": [
   "Transformer_(architecture)"
  ],
  "Post-training": [
   "Instruction_tuning",
   "Reinforcement_learning_from_human_feedback"
  ],
  "Paper/report": [
   "OpenAI",
   "GPT-4o"
  ]
 },
 "aliases": [],
 "family": "GPT (model family)",
 "org": "OpenAI",
 "registry_status": "stub",
 "words": 341,
 "references": 3,
 "outbound": [
  "Claude_3.5_Haiku",
  "GPT-3.5",
  "GPT-4_Turbo",
  "GPT-4o",
  "GPT-5",
  "GPT_(model_family)",
  "Gemini_1.5",
  "Instruction_tuning",
  "Knowledge_distillation",
  "Large_language_model",
  "MMLU",
  "OpenAI",
  "Reinforcement_learning_from_human_feedback",
  "Transformer_(architecture)",
  "o1",
  "o4-mini"
 ],
 "inbound": [
  "Claude_3.5_Haiku",
  "GPT-4.5",
  "GPT-4_Turbo",
  "GPT-5.2",
  "GPT-5.6",
  "o4-mini"
 ],
 "url": "wiki/GPT-4o_mini.html",
 "built": "2026-07-24"
}