metadata

license: apache-2.0
tags:
  - mlx

voxmenthe/Yi-1.5-34B-Chat-16K-8bit

The Model voxmenthe/Yi-1.5-34B-Chat-16K-8bit was converted to MLX format from 01-ai/Yi-1.5-34B-Chat-16K using mlx-lm version 0.13.1.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("voxmenthe/Yi-1.5-34B-Chat-16K-8bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)