Instructions to use vince62s/phi-2-psy with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vince62s/phi-2-psy with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="vince62s/phi-2-psy", trust_remote_code=True)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("vince62s/phi-2-psy", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("vince62s/phi-2-psy", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use vince62s/phi-2-psy with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "vince62s/phi-2-psy" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "vince62s/phi-2-psy", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/vince62s/phi-2-psy
- SGLang
How to use vince62s/phi-2-psy 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 "vince62s/phi-2-psy" \ --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": "vince62s/phi-2-psy", "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 "vince62s/phi-2-psy" \ --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": "vince62s/phi-2-psy", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use vince62s/phi-2-psy with Docker Model Runner:
docker model run hf.co/vince62s/phi-2-psy
how to consume the output merge model?
I see that you guys have modified the mergekit and added the custom inference code. After trying to modify the mergekit-moe, I finally have a merged phi2 model, but honestly no clue on how to consume it, I am going through the configuration_phi.py and modeling_phi.py from your repo, probably I cannot use them as such because of the way I modified the mergekit-moe or is that irrelevant?
cause now running the inference script while using your configuration_phi.py and modeling.py gives me an error with accelerate lib as follows. Could you please shed some light? @vince62s