---
title: "POST /api/v1/public/finetuning/create"
method: POST
path: "/api/v1/public/finetuning/create"
tags: ["Public Fine-tuning"]
---

# POST /api/v1/public/finetuning/create

`POST /api/v1/public/finetuning/create`

Create and start fine-tuning job

## Request body

- CreateFinetuningRequest
  - `snapshot_id` string, uuid, required
  - `name` string, required
  - `experiments` object[], required
    - `base_model_id` string, required
    - `batch_size` integer, required
    - `learning_rate_multiplier` number, required
    - `n_epochs` integer, required
    - `training_type` 'full' | 'lora' | 'qlora', required
    - `refinetune_from_experiment_id` string, uuid

## Response `200`

Fine-tuning job created successfully

- FinetuningResponse
  - `job_id` string, uuid, required

## Other responses

- `400` — Bad request
- `402` — Insufficient credits
- `404` — Snapshot not found
- `429` — Quota exceeded
- `500` — Internal server error

## Changes

- **2025-12-12** `2cb1f5df1b89` — 1 breaking, 1 warning, 1 info
  - added the new required request property `experiments/items/training_type`
  - removed the request property `experiments/items/lora`
  - added the new optional request property `experiments/items/refinetune_from_experiment_id`
- **2025-11-12** `25d63a01824a` — 1 info
  - endpoint added

[Change history](https://skmtc.dev/premai-io/apis/prem-studio-api/changes/api/v1/public/finetuning/create/post.md)

---

[API](https://skmtc.dev/premai-io/apis/prem-studio-api.md) · [All operations](https://skmtc.dev/premai-io/apis/prem-studio-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/premai-io/prem-studio-api/revisions/2cb1f5df1b89/schema)