Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
Python agentaichatgptdeepseekfine-tuninggemmaimage-generationllamallmllmsopenaipythonqwenreinforcement-learningself-hostedstable-diffusiontext-to-speechttsuiunsloth
# SPDX-License-Identifier: AGPL-3.0-only
# Copyright 2026-present the Unsloth AI Inc. team. All rights reserved. See /studio/LICENSE.AGPL-3.0
from __future__ import annotations
import sys
from pathlib import Path
from types import SimpleNamespace
import typer
from typer.testing import CliRunner
_REPO_ROOT = Path(__file__).resolve().parents[2]
if str(_REPO_ROOT) not in sys.path:
sys.path.insert(0, str(_REPO_ROOT))
import unsloth_cli.commands.train as train_module # noqa: E402
from unsloth_cli.config import Config # noqa: E402
class _RecordingTrainer:
is_vlm = False
training_thread = None
def __init__(self):
self.prepare_kwargs = None
def load_model(self, **kwargs):
return True
def prepare_model_for_training(self, **kwargs):
self.prepare_kwargs = kwargs
return True
def load_and_format_dataset(self, **kwargs):
return [], None
def start_training(self, **kwargs):
return True
def get_training_progress(self):
return SimpleNamespace(error = None, is_completed = True)
def test_random_seed_flag_seeds_the_lora_init(tmp_path, monkeypatch):
config = tmp_path / "config.yaml"
config.write_text(
"model: unsloth/Qwen3-0.6B\ndata:\n dataset: yahma/alpaca-cleaned\n",
encoding = "utf-8",
)
trainer = _RecordingTrainer()
monkeypatch.setattr(train_module, "_create_cli_trainer", lambda *args: trainer)
monkeypatch.setattr(
train_module, "_dataset_modalities", lambda *args: (False, False), raising = False
)
app = typer.Typer()
app.command()(train_module.train)
result = CliRunner().invoke(app, ["--config", str(config), "--random-seed", "42"])
assert result.exit_code == 0, result.output
assert trainer.prepare_kwargs["random_state"] == 42
def test_mlx_adapter_accepts_the_cli_lora_kwargs():
backend = str(_REPO_ROOT / "studio" / "backend")
if backend not in sys.path:
sys.path.insert(0, backend)
from core.training.training import create_mlx_trainer_adapter
adapter = create_mlx_trainer_adapter()
adapter._model_config = {}
assert adapter.prepare_model_for_training(**Config().model_kwargs(True, False))