CoolFace
Modelpublic

second-state/EXAONE-Deep-32B-GGUF

sourceHugging Faceotherupdated 2y agoView on Hugging Face
0likes734downloads
Model Card

<!-- header start --> <!-- 200823 --> <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;"> <!-- header end -->

EXAONE-Deep-32B-GGUF

Original Model

LGAI-EXAONE/EXAONE-Deep-32B

Run with LlamaEdge

  • —LlamaEdge version: v0.16.12 and above
  • —Prompt template
  • —Prompt type: exaone-deep-chat
  • —Prompt string
text
    [|system|]{system_message}[|endofturn|]
    [|user|]{user_message_1}
    [|assistant|]{assistant_message_1}[|endofturn|]
    [|user|]{user_message_2}
    [|assistant|]<thought>
  • —Context size: 32000
  • —Run as LlamaEdge service
bash
  wasmedge --dir .:. --nn-preload default:GGML:AUTO:EXAONE-Deep-32B-Q5_K_M.gguf \
      llama-api-server.wasm \
      --prompt-template exaone-deep-chat \
      --ctx-size 32000 \
      --model-name EXAONE-Deep-32B
  • —Run as LlamaEdge command app
bash
  wasmedge --dir .:. --nn-preload default:GGML:AUTO:EXAONE-Deep-32B-Q5_K_M.gguf \
    llama-chat.wasm \
    --prompt-template exaone-deep-chat \
    --ctx-size 32000

Quantized GGUF Models

NameQuant methodBitsSizeUse case
EXAONE-Deep-32B-Q2_K.ggufQ2_K211.9 GBsmallest, significant quality loss - not recommended for most purposes
EXAONE-Deep-32B-Q3_K_L.ggufQ3KL316.8 GBsmall, substantial quality loss
EXAONE-Deep-32B-Q3_K_M.ggufQ3KM315.5 GBvery small, high quality loss
EXAONE-Deep-32B-Q3_K_S.ggufQ3KS314.0 GBvery small, high quality loss
EXAONE-Deep-32B-Q4_0.ggufQ4_0418.1 GBlegacy; small, very high quality loss - prefer using Q3KM
EXAONE-Deep-32B-Q4_K_M.ggufQ4KM419.3 GBmedium, balanced quality - recommended
EXAONE-Deep-32B-Q4_K_S.ggufQ4KS418.3 GBsmall, greater quality loss
EXAONE-Deep-32B-Q5_0.ggufQ5_0522.1 GBlegacy; medium, balanced quality - prefer using Q4KM
EXAONE-Deep-32B-Q5_K_M.ggufQ5KM522.7 GBlarge, very low quality loss - recommended
EXAONE-Deep-32B-Q5_K_S.ggufQ5KS522.1 GBlarge, low quality loss - recommended
EXAONE-Deep-32B-Q6_K.ggufQ6_K626.3 GBvery large, extremely low quality loss
EXAONE-Deep-32B-Q8_0.ggufQ8_0834.0 GBvery large, extremely low quality loss - not recommended
EXAONE-Deep-32B-f16-00001-of-00003.gguff161629.8 GB
EXAONE-Deep-32B-f16-00002-of-00003.gguff161630.0 GB
EXAONE-Deep-32B-f16-00003-of-00003.gguff16164.23 GB

Quantized with llama.cpp b4914.