README.md

metadata

tags:
  - swe-bench
  - mlx
license: apache-2.0
metrics:
  - pass@1
library_name: mlx
pipeline_tag: text-generation
base_model: Skywork/Skywork-SWE-32B

mlx-community/Skywork-SWE-32B-8bit

This model mlx-community/Skywork-SWE-32B-8bit was converted to MLX format from Skywork/Skywork-SWE-32B 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/Skywork-SWE-32B-8bit")

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)