README.md

metadata

language:
  - en
  - ja
library_name: mlx
pipeline_tag: text-generation
license:
  - llama3.3
  - gemma
model_type: llama
datasets:
  - tokyotech-llm/lmsys-chat-1m-synth
  - lmsys/lmsys-chat-1m
base_model: tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.5
tags:
  - mlx

mlx-community/Llama-3.1-Swallow-8B-Instruct-v0.5-4bit

This model mlx-community/Llama-3.1-Swallow-8B-Instruct-v0.5-4bit was converted to MLX format from tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.5 using mlx-lm version 0.25.2.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("mlx-community/Llama-3.1-Swallow-8B-Instruct-v0.5-4bit")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)