metadata
license: mit
datasets:
- allenai/WildChat-1M
language:
- en
base_model: microsoft/UserLM-8b
pipeline_tag: text-generation
tags:
- userlm
- simulation
- mlx
library_name: mlx
Otilde/UserLM-8b-MLX-Mixed-4_6
This model Otilde/UserLM-8b-MLX-Mixed-4_6 was converted to MLX format from microsoft/UserLM-8b using mlx-lm version 0.28.2.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("Otilde/UserLM-8b-MLX-Mixed-4_6")
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)