Instructions to use lmsys/vicuna-7b-v1.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lmsys/vicuna-7b-v1.5 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="lmsys/vicuna-7b-v1.5")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("lmsys/vicuna-7b-v1.5") model = AutoModelForCausalLM.from_pretrained("lmsys/vicuna-7b-v1.5", device_map="auto") - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use lmsys/vicuna-7b-v1.5 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "lmsys/vicuna-7b-v1.5" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "lmsys/vicuna-7b-v1.5", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/lmsys/vicuna-7b-v1.5
- SGLang
How to use lmsys/vicuna-7b-v1.5 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 "lmsys/vicuna-7b-v1.5" \ --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": "lmsys/vicuna-7b-v1.5", "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 "lmsys/vicuna-7b-v1.5" \ --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": "lmsys/vicuna-7b-v1.5", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use lmsys/vicuna-7b-v1.5 with Docker Model Runner:
docker model run hf.co/lmsys/vicuna-7b-v1.5
Instruction format?
Hello!
I have been experimenting with this model and i seem to be having some problem with the formats used.
At first I tried using the original vicuna format:
### Human:
Question
### Assistant:
Answer
And the outputs i got seemed strange, for example:
### Human: How many languages do you speak ### Assistant: 1. nobody
After that i tried using the vicuna 1.1 format:
USER:
Question
ASSISTANT:
Answer
And still got strange results. I have tried adding and removing the system prompt, but that didnt seem to work.
¿Is the instruction different from what i expected?
Did you find a solution ?
No, I convinced myself that the model was not as reliable as I thought, and moved on to another. Sorry I can't be of help.