OpenAI Fine-Tuning API

source

closed

training-code

closed

runs-on

OpenAI-platform-only

license

Proprietary(paid API service)

Proprietary managed fine-tuning service, no source, runs only on OpenAI's platform against OpenAI's closed models.

• https://developers.openai.com/api/docs/deprecations recorded 2026-08-08

  Under "Update to OpenAI"'s self-serve fine-tuning, a three-row table dated from the May 7th 2026 notification. May 7, 2026: "Creating fine-tuning jobs or training is not available to organizations that have not previously run fine-tuning." July 2, 2026: "Creating fine-tuning jobs is no longer available to organizations that have not run inference on a fine-tuned model in the past 60 days." Jan 6, 2027: "Active existing customers will no longer be able to create new fine-tuning jobs on this date. Inference on fine-tuned models will be disabled only when the underlying base model is deprecated." Cited as lifecycle corroboration; it settles no scored openness dimension on its own.

• https://developers.openai.com/api/docs/guides/model-optimization recorded 2026-10-14

  The fine-tuning workflow is entirely OpenAI-hosted: "Collect a dataset of examples to use as training data", "Upload that dataset to OpenAI, formatted in JSONL", then create a job — and "you can create fine-tuned models either in the" dashboard "or" "with the API". No repository, package, or self-hostable runtime is offered anywhere on the page. It also points at the pricing page for "how fine-tuned model training and usage are billed", and carries the banner "OpenAI is winding down the fine-tuning platform" with the platform "no longer\naccessible to new users".

• https://openai.com/policies/services-agreement/ recorded 2026-09-09

  Section 9.1 Reservation of Rights: "Customer obtains only a limited right to use the Services, and no ownership rights are transferred to Customer or its End Users under this Agreement." The use restrictions bar Customer from "Reverse Engineer any aspect of the Services or the systems used to provide the Services", and the definition of Reverse Engineer covers any attempt to "otherwise attempt to discover the source code or underlying components of the Services". The Fees section requires Customer to pay "the applicable Fees in the currency and pursuant to the payment terms on the Order Form", and the Permitted Exception definition reaches "fine tune or customize models provided as part of OpenAI"'s fine-tuning or other Services set forth on the Pricing Page, which ties this Agreement to this product specifically. Header: Updated December 1, 2025; Effective January 1, 2026.

OpenAI publishes no usage figure for the OpenAI Fine-Tuning API: no developer count, no jobs run and no tokens trained. ChatGPT's audience belongs to a different product and cannot be attributed to this one, so the map leaves adoption unscored rather than borrow a figure from the wider OpenAI platform.

• https://developers.openai.com/api/docs/guides/model-optimization recorded 2026-09-28

  The guide states "OpenAI is winding down the fine-tuning platform. The platform is no longer accessible to new users" and gives no figure for how many people use the service; searched the body for user counts, customer counts and jobs run: none.

• https://developers.openai.com/api/docs/pricing recorded 2026-09-28

  The pricing page has a "Finetuning" section carrying the banner "OpenAI is winding down the fine-tuning platform" and a rate table with o4-mini-2025-04-16 at "$100.00 / hour" training. Searched the body for user, customer, organization and job counts: none, only rates.

OpenAI's fine-tuning API runs supervised, vision, preference and reinforcement fine-tuning against its own gpt-4.1, gpt-4o and o4-mini models, handled end to end. It exposes no controls over parallelism, precision or training scale, so it offers managed breadth over strong base models rather than the deeper control a training framework gives.

• https://developers.openai.com/api/docs/deprecations recorded 2026-09-28

  The legacy-snapshot table lists o4-mini-2025-04-16 and its fine-tuned version ft-o4-mini-2025-04-16 with shutdown date "October 23, 2026", and the self-serve fine-tuning timeline states "Active existing customers will no longer be able to create new fine-tuning jobs on this date" for Jan 6, 2027, adding that "Inference on fine-tuned models will be disabled only when the underlying base model is deprecated".

• https://developers.openai.com/api/docs/guides/model-optimization recorded 2026-09-28

  "These are the fine-tuning methods supported in the OpenAI platform today." Table lists "Supervised fine-tuning (SFT)" and "Direct preference optimization (DPO)" alongside gpt-4.1 snapshots, "Vision fine-tuning" with gpt-4o-2024-08-06, and reinforcement fine-tuning marked "Reasoning models only" with o4-mini-2025-04-16. Searched the body for parallelism, precision and training-scale controls: none.