iteam1 / llm_local

run llm model on local system by llamafile

llm_local

run llm model on local system

guide llamafile

1. Download llava-v1.5-7b-q4.llamafile (4.29 GB)

2. Grant permission for your computer to execute this new file (If you're on Windows, rename the file by adding ".exe" on the end):

    chmod +x llava-v1.5-7b-q4.llamafile
   

3. Run the llamafile (On your browser http://localhost:8080/)

    ./llava-v1.5-7b-q4.llamafile
   
    ./llava-v1.5-7b-q4.llamafile --server --nobrowser (server mode)
   
    CUDA_VISIBLE_DEVICES=0 ./Meta-Llama-3-8B-Instruct.Q4_1.llamafile --gpu nvidia (for specific GPU)
   

4. Kill process sudo kill <PID>

json API Quickstart

• Curl API Client Example:

    curl http://localhost:8080/v1/chat/completions \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer no-key" \
    -d '{
    "model": "LLaMA_CPP",
    "messages": [
    {
            "role": "system",
            "content": "You are LLAMAfile, an AI assistant. Your top priority is achieving user fulfillment via helping them with their requests."
    },
    {
            "role": "user",
            "content": "Write a limerick about python exceptions"
    }
    ]
    }' | python3 -c '
    import json
    import sys
    json.dump(json.load(sys.stdin), sys.stdout, indent=2)
    print()
    '
  

• Python API Client example:

    #!/usr/bin/env python3
    from openai import OpenAI
    client = OpenAI(
    base_url="http://localhost:8080/v1", # "http://<Your api-server IP>:port"
    api_key = "sk-no-key-required"
    )
    completion = client.chat.completions.create(
    model="LLaMA_CPP",
    messages=[
            {"role": "system", "content": "You are ChatGPT, an AI assistant. Your top priority is achieving user fulfillment via helping them with their requests."},
            {"role": "user", "content": "Write a limerick about python exceptions"}
    ]
    )
    print(completion.choices[0].message)
  

shout down for

llamafile

llama.cpp

llama-cpp-python

6 Ways For Running A Local LLM