mlx-examples
mlx-examples copied to clipboard
Examples in the MLX framework
Hello, thanks for the amazing repo. I would like to request support for function calling feature for the mlx_lm server, similar to OpenAI's implementation. Please let me know if this...
```python from mlx_lm import load, generate model, tokenizer = load(path_or_hf_repo = phi3_path,#,tokenizer_config=tokenizer_config # adapter_path = adapt_path, tokenizer_config={"trust_remote_code": True}) messages = [ {"role": "user", "content": 'Where is delhi, explain it only...
Per the title, allow a structured `hf_dataset` YAML configuration parameter for specifying an HF hub-hosted dataset (via `name`) to use with training and the ability to use [datasets](https://huggingface.co/docs/datasets/en/index)' local file...
This PR adds support for openELM. Twitter: @Prince_Canuma Todo: - [x] 3-8B - [x] 20-34B
Added proper error handling to `load` and `get_model_path` by adding a dedicated exception class, because when the local path is not right, it still throws the huggingface `RepositoryNotFoundError`, which is...
Add API generation documentation for when user wants to generate using fine tuned adapters.
The speed benefits over whisper.cpp on Apple GPU are amazing, but I believe only text / segments / words output format is currently supported. It would be amazing if the...
Using ```python import mlx_whisper speech_file = "stereo.mp3" text = mlx_whisper.transcribe(speech_file,path_or_hf_repo=f"mlx-community/whisper-large-v3-mlx",verbose=False)["text"] f=open("result.txt","w+") f.write(text) f.close() ``` I find that the output contains repeated phrases from time to time, enough to ruin the...
Hi, When I train my model using the code below (Note this was all done in jupyter Notebook). ``` !python -m mlx_lm.lora \ --model mlx-community/Meta-Llama-3-8B-Instruct-4bit \ --train \ --batch-size 1...
There are MLX models for text generation (llama 3) and for text recognition (whisper) but I think that to have a complete NLP environment it would be necessary to create...