README.md

metadata

datasets:
  - openbmb/UltraFeedback
language:
  - en
license: apache-2.0
pipeline_tag: text-generation
tags:
  - mlx

voxmenthe/Llama-3-Instruct-8B-SPPO-Iter3-fp16

The Model voxmenthe/Llama-3-Instruct-8B-SPPO-Iter3-fp16 was converted to MLX format from UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3 using mlx-lm version 0.15.0.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("voxmenthe/Llama-3-Instruct-8B-SPPO-Iter3-fp16")
response = generate(model, tokenizer, prompt="hello", verbose=True)