metadata
license: apache-2.0
datasets:
- Tongyi-Zhiwen/DocQA-RL-1.6K
base_model: Tongyi-Zhiwen/QwenLong-L1-32B
tags:
- long-context
- large-reasoning-model
- mlx
- mlx-my-repo
WaveCut/QwenLong-L1-32B-mlx-8Bit
The Model WaveCut/QwenLong-L1-32B-mlx-8Bit was converted to MLX format from Tongyi-Zhiwen/QwenLong-L1-32B using mlx-lm version 0.22.3.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("WaveCut/QwenLong-L1-32B-mlx-8Bit")
prompt="hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)