README.md

metadata

license: apache-2.0
base_models:
  - open-r1/OlympicCoder-7B
  - Qwen/Qwen2.5-Coder-7B-Instruct
  - CodeV-R1-Qwen-7B
  - TIGER-Lab/VisCoder-7B
language:
  - en
pipeline_tag: text-generation
tags:
  - merge
  - programming
  - code generation
  - code
  - codeqwen
  - moe
  - coding
  - coder
  - qwen2
  - chat
  - qwen
  - qwen-coder
  - mixture of experts
  - qwen2moe
  - 4X7B
  - shared expert
  - mlx
library_name: mlx
base_model: DavidAU/Qwen2.5-4x7B-Quad-Coder-Instruct-30B

Qwen2.5-4x7B-Quad-Coder-Instruct-30B-q8-mlx

This model Qwen2.5-4x7B-Quad-Coder-Instruct-30B-q8-mlx was converted to MLX format from DavidAU/Qwen2.5-4x7B-Quad-Coder-Instruct-30B using mlx-lm version 0.26.0.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("Qwen2.5-4x7B-Quad-Coder-Instruct-30B-q8-mlx")

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)