Instructions to use m-a-p/ChatMusician with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use m-a-p/ChatMusician with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="m-a-p/ChatMusician")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("m-a-p/ChatMusician") model = AutoModelForCausalLM.from_pretrained("m-a-p/ChatMusician") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use m-a-p/ChatMusician with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "m-a-p/ChatMusician" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "m-a-p/ChatMusician", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/m-a-p/ChatMusician
- SGLang
How to use m-a-p/ChatMusician with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "m-a-p/ChatMusician" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "m-a-p/ChatMusician", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "m-a-p/ChatMusician" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "m-a-p/ChatMusician", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use m-a-p/ChatMusician with Docker Model Runner:
docker model run hf.co/m-a-p/ChatMusician
Small bugs in the dataset
#4
by t1u1 - opened
From Figure 9 of the paper:
Which of the following best describes the seventh chord in the above example?
L:1/4
M:4/4
K:D
[FGBd]4 |] %1
A. Major seventh in third inversion
B. Dominant seventh in second inversion
C. Major/minor seventh in third inversion
D. Minor seventh in second inversion
Answer: A
Here, the word "above" is wrong, because the example is below not above. All the examples in that figure have the same bug.
I wonder whether the LLM gets confused or learns wrong meanings by such bugs.
Thanks a lot for reminding us about the bug!
This bug is introduced in our evaluation template code (will be open-sourced in our github repo soon).
We are planning a new test soon. We will rearrange the template and see how much it affects the result.
New version will look something like this:
L:1/4
M:4/4
K:D
[FGBd]4 |] %1
Which of the following best describes the seventh chord in the above example?
A. Major seventh in third inversion
B. Dominant seventh in second inversion
C. Major/minor seventh in third inversion
D. Minor seventh in second inversion
Answer: A