QuantFactory/Meta-Llama-3-120B-Instruct-GGUF

Meta-Llama-3-120B-Instruct- GGUF

• This is quantized version of mlabonne/Meta-Llama-3-120B-Instruct created using llama.cpp

Model Description

Meta-Llama-3-120B-Instruct is a self-merge with meta-llama/Meta-Llama-3-70B-Instruct.

It was inspired by large merges like alpindale/goliath-120b, nsfwthrowitaway69/Venus-120b-v1.0, cognitivecomputations/MegaDolphin-120b, and wolfram/miquliz-120b-v2.0.

No eval yet, but it is approved by Eric Hartford: https://twitter.com/erhartford/status/1787050962114207886

🧩 Configuration

slices:
- sources:
  - layer_range: [0, 20]
    model: meta-llama/Meta-Llama-3-70B-Instruct
- sources:
  - layer_range: [10, 30]
    model: meta-llama/Meta-Llama-3-70B-Instruct
- sources:
  - layer_range: [20, 40]
    model: meta-llama/Meta-Llama-3-70B-Instruct
- sources:
  - layer_range: [30, 50]
    model: meta-llama/Meta-Llama-3-70B-Instruct
- sources:
  - layer_range: [40, 60]
    model: meta-llama/Meta-Llama-3-70B-Instruct
- sources:
  - layer_range: [50, 70]
    model: meta-llama/Meta-Llama-3-70B-Instruct
- sources:
  - layer_range: [60, 80]
    model: meta-llama/Meta-Llama-3-70B-Instruct
merge_method: passthrough
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/Llama-3-120B"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])