Instructions to use Ailiance-fr/devstral-python-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Ailiance-fr/devstral-python-lora with PEFT:
Task type is invalid.
- MLX
How to use Ailiance-fr/devstral-python-lora with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("Ailiance-fr/devstral-python-lora") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
- MLX LM
How to use Ailiance-fr/devstral-python-lora with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "Ailiance-fr/devstral-python-lora" --prompt "Once upon a time"
| { | |
| "adapter_path": "/Users/clems/KIKI-Mac_tunner/output/eu-kiki/devstral-python", | |
| "batch_size": 1, | |
| "clear_cache_threshold": 0, | |
| "config": "/Users/clems/KIKI-Mac_tunner/output/eu-kiki/devstral-python/train_config.yaml", | |
| "data": "/Users/clems/KIKI-Mac_tunner/data/micro-kiki/python", | |
| "fine_tune_type": "lora", | |
| "grad_accumulation_steps": 4, | |
| "grad_checkpoint": true, | |
| "iters": 500, | |
| "learning_rate": 1e-05, | |
| "lora_parameters": { | |
| "alpha": 32, | |
| "dropout": 0.05, | |
| "rank": 16, | |
| "scale": 2.0 | |
| }, | |
| "lr_schedule": null, | |
| "mask_prompt": false, | |
| "max_seq_length": 2048, | |
| "model": "/Users/clems/KIKI-Mac_tunner/models/Devstral-Small-2-24B-Instruct-2512", | |
| "num_layers": -1, | |
| "optimizer": "adam", | |
| "optimizer_config": { | |
| "adam": {}, | |
| "adamw": {}, | |
| "muon": {}, | |
| "sgd": {}, | |
| "adafactor": {} | |
| }, | |
| "project_name": null, | |
| "report_to": null, | |
| "resume_adapter_file": null, | |
| "save_every": 100, | |
| "seed": 42, | |
| "steps_per_eval": 100, | |
| "steps_per_report": 5, | |
| "test": false, | |
| "test_batches": 500, | |
| "train": true, | |
| "val_batches": 10 | |
| } |