README.md

metadata

license: llama3
language:
  - en
tags:
  - finance
  - mlx
  - mlx-my-repo
datasets:
  - Open-Orca/OpenOrca
  - GAIR/lima
  - WizardLM/WizardLM_evol_instruct_V2_196k
base_model: instruction-pretrain/finance-Llama3-8B

codingmavin/finance-Llama3-8B-mlx-6Bit

The Model codingmavin/finance-Llama3-8B-mlx-6Bit was converted to MLX format from instruction-pretrain/finance-Llama3-8B using mlx-lm version 0.22.1.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("codingmavin/finance-Llama3-8B-mlx-6Bit")

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)