metadata
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-14B-Instruct-1M/blob/main/LICENSE
language:
- en
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-14B-Instruct-1M
tags:
- chat
- mlx
- mlx-my-repo
library_name: transformers
valuat/Qwen2.5-14B-Instruct-1M-mlx-fp16
The Model valuat/Qwen2.5-14B-Instruct-1M-mlx-fp16 was converted to MLX format from Qwen/Qwen2.5-14B-Instruct-1M using mlx-lm version 0.26.4.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("valuat/Qwen2.5-14B-Instruct-1M-mlx-fp16")
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)