Modelpublic

second-state/llm-compiler-13b-GGUF

sourceHugging Faceotherupdated 2y agoView on Hugging Face

0likes568downloads

Model Card

<div style="width: auto; margin-left: auto; margin-right: auto"> <img src="https://github.com/LlamaEdge/LlamaEdge/raw/dev/assets/logo.svg" style="width: 100%; min-width: 400px; display: block; margin: auto;"> </div> <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">

llm-compiler-13b-GGUF

Original Model

facebook/llm-compiler-13b

License A custom commercial license is available at: https://ai.meta.com/resources/models-and-libraries/llama-downloads/

Run with LlamaEdge

—LlamaEdge version: coming soon

<!-- - Prompt template

—Prompt type: codellama-instruct

—Prompt string

text

    [INST] {system_message}

    <code>
    {user_message}
    </code>[/INST]

Note that the system_message should be Tell me how to optimize this LLVM-IR for object file size: or Disassemble this code to LLVM-IR:.

—Context size: 4096

—Run as LlamaEdge command app

bash

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:llm-compiler-13b-Q5_K_M.gguf llama-chat.wasm -p codellama-instruct

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
llm-compiler-13b-Q2_K.gguf	Q2_K	2	4.85 GB	smallest, significant quality loss - not recommended for most purposes
llm-compiler-13b-Q3_K_L.gguf	Q3KL	3	6.93 GB	small, substantial quality loss
llm-compiler-13b-Q3_K_M.gguf	Q3KM	3	6.34 GB	very small, high quality loss
llm-compiler-13b-Q3_K_S.gguf	Q3KS	3	5.66 GB	very small, high quality loss
llm-compiler-13b-Q4_0.gguf	Q4_0	4	7.37 GB	legacy; small, very high quality loss - prefer using Q3KM
llm-compiler-13b-Q4_K_M.gguf	Q4KM	4	7.87 GB	medium, balanced quality - recommended
llm-compiler-13b-Q4_K_S.gguf	Q4KS	4	7.42 GB	small, greater quality loss
llm-compiler-13b-Q5_0.gguf	Q5_0	5	8.97 GB	legacy; medium, balanced quality - prefer using Q4KM
llm-compiler-13b-Q5_K_M.gguf	Q5KM	5	9.23 GB	large, very low quality loss - recommended
llm-compiler-13b-Q5_K_S.gguf	Q5KS	5	8.97 GB	large, low quality loss - recommended
llm-compiler-13b-Q6_K.gguf	Q6_K	6	10.7 GB	very large, extremely low quality loss
llm-compiler-13b-Q8_0.gguf	Q8_0	8	13.8 GB	very large, extremely low quality loss - not recommended
llm-compiler-13b-f16.gguf	f16	f16	26 GB

Quantized with llama.cpp b3259