Fine Tuning

+++ disableToc = false title = "Fine-Tuning" weight = 18 url = '/features/fine-tuning/' +++

LocalAI supports fine-tuning LLMs directly through the API and Web UI. Fine-tuning is powered by pluggable backends that implement a generic gRPC interface, allowing support for different training frameworks and model types.

Supported Backends

Availability

Fine-tuning is always enabled. When authentication is enabled, fine-tuning is a per-user feature (default OFF). Admins can enable it for specific users via the user management API.

{{% notice note %}} This feature is experimental and may change in future releases. {{% /notice %}}

Quick Start

1. Start a fine-tuning job

curl -X POST http://localhost:8080/api/fine-tuning/jobs \
  -H "Content-Type: application/json" \
  -d '{
    "model": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
    "backend": "trl",
    "training_method": "sft",
    "training_type": "lora",
    "dataset_source": "yahma/alpaca-cleaned",
    "num_epochs": 1,
    "batch_size": 2,
    "learning_rate": 0.0002,
    "adapter_rank": 16,
    "adapter_alpha": 16,
    "extra_options": {
      "max_seq_length": "512"
    }
  }'

2. Monitor progress (SSE stream)

curl -N http://localhost:8080/api/fine-tuning/jobs/{job_id}/progress

3. List checkpoints

curl http://localhost:8080/api/fine-tuning/jobs/{job_id}/checkpoints

4. Export model

curl -X POST http://localhost:8080/api/fine-tuning/jobs/{job_id}/export \
  -H "Content-Type: application/json" \
  -d '{
    "export_format": "gguf",
    "quantization_method": "q4_k_m",
    "output_path": "/models/my-finetuned-model"
  }'

API Reference

Endpoints

Job Request Fields

Backend-Specific Options (extra_options)

TRL

DPO-specific (training_method=dpo)

GRPO-specific (training_method=grpo)

GRPO Reward Functions

GRPO training requires reward functions to evaluate model completions. Specify them via the reward_functions field (a typed array) or via extra_options["reward_funcs"] (a JSON string).

Built-in Reward Functions

Inline Custom Reward Functions

You can provide custom reward function code as a Python function body. The function receives completions (list of strings) and **kwargs, and must return list[float].

Security restrictions for inline code:

• Allowed builtins: len, int, float, str, list, dict, range, enumerate, zip, map, filter, sorted, min, max, sum, abs, round, any, all, isinstance, print, True, False, None
• Available modules: re, math, json, string
• Blocked: open, __import__, exec, eval, compile, os, subprocess, getattr, setattr, delattr, globals, locals
• Functions are compiled and validated at job start (fail-fast on syntax errors)

Example API Request

curl -X POST http://localhost:8080/api/fine-tuning/jobs \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Qwen/Qwen2.5-1.5B-Instruct",
    "backend": "trl",
    "training_method": "grpo",
    "training_type": "lora",
    "dataset_source": "my-reasoning-dataset",
    "num_epochs": 1,
    "batch_size": 2,
    "learning_rate": 5e-6,
    "reward_functions": [
      {"type": "builtin", "name": "reasoning_accuracy_reward"},
      {"type": "builtin", "name": "format_reward"},
      {"type": "builtin", "name": "length_reward", "params": {"target_length": "200"}},
      {"type": "inline", "name": "think_presence", "code": "return [1.0 if \"<think>\" in c else 0.0 for c in completions]"}
    ],
    "extra_options": {
      "num_generations": "4",
      "max_completion_length": "256"
    }
  }'

Export Formats

GGUF Quantization Methods

q4_k_m, q5_k_m, q8_0, f16, q4_0, q5_0

Web UI

When fine-tuning is enabled, a "Fine-Tune" page appears in the sidebar under the Agents section. The UI provides:

1. Job Configuration — Select backend, model, training method, adapter type, and hyperparameters
2. Dataset Upload — Upload local datasets or reference HuggingFace datasets
3. Training Monitor — Real-time loss chart, progress bar, metrics display
4. Export — Export trained models in various formats

Dataset Formats

Datasets should follow standard HuggingFace formats:

• SFT: Alpaca format (instruction, input, output fields) or ChatML/ShareGPT
• DPO: Preference pairs (prompt, chosen, rejected fields)
• GRPO: Prompts with reward signals

Supported file formats: .json, .jsonl, .csv

Architecture

Fine-tuning uses the same gRPC backend architecture as inference:

1. Proto layer: FineTuneRequest, FineTuneProgress (streaming), StopFineTune, ListCheckpoints, ExportModel
2. Python backends: Each backend implements the gRPC interface with its specific training framework
3. Go service: Manages job lifecycle, routes API requests to backends
4. REST API: HTTP endpoints with SSE progress streaming
5. React UI: Configuration form, real-time training monitor, export panel