README.md

metadata

license: llama3.1
model_name: Meta-Llama-3.1-70B-Instruct
base_model: meta-llama/Meta-Llama-3.1-70B-Instruct
inference: false
model_creator: Meta Llama3
model_type: llama
pipeline_tag: text-generation
quantized_by: Second State Inc.
language:
  - en
  - de
  - fr
  - it
  - pt
  - hi
  - es
  - th
tags:
  - facebook
  - meta
  - pytorch
  - llama
  - llama-3

---

Meta-Llama-3.1-70B-Instruct-GGUF

Original Model

meta-llama/Meta-Llama-3.1-70B-Instruct

Run with LlamaEdge

• LlamaEdge version: v0.12.4 and above

• Prompt template

  • Prompt type: llama-3-chat

  • Prompt string

    <|begin_of_text|><|start_header_id|>system<|end_header_id|>
    
    {{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>
    
    {{ user_message_1 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
    
    {{ model_answer_1 }}<|eot_id|><|start_header_id|>user<|end_header_id|>
    
    {{ user_message_2 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
    

• Context size: 128000

• Run as LlamaEdge service

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:Meta-Llama-3.1-70B-Instruct-Q4_K_M.gguf \
    llama-api-server.wasm \
    --prompt-template llama-3-chat \
    --ctx-size 128000 \
    --model-name Llama-3.1-70b
  

• Run as LlamaEdge command app

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:Meta-Llama-3.1-70B-Instruct-Q4_K_M.gguf \
    llama-chat.wasm \
    --prompt-template llama-3-chat \
    --ctx-size 128000
  

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
Meta-Llama-3.1-70B-Instruct-Q2_K.gguf	Q2_K	2	26.4 GB	smallest, significant quality loss - not recommended for most purposes
Meta-Llama-3.1-70B-Instruct-Q3_K_L.gguf	Q3_K_L	3	37.1 GB	small, substantial quality loss
Meta-Llama-3.1-70B-Instruct-Q3_K_M.gguf	Q3_K_M	3	34.3 GB	very small, high quality loss
Meta-Llama-3.1-70B-Instruct-Q3_K_S.gguf	Q3_K_S	3	30.9 GB	very small, high quality loss
Meta-Llama-3.1-70B-Instruct-Q4_0.gguf	Q4_0	4	40 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Meta-Llama-3.1-70B-Instruct-Q4_K_M.gguf	Q4_K_M	4	42.5 GB	medium, balanced quality - recommended

Quantized with llama.cpp 3445.