CoolFace
Modelpublic

second-state/E5-Mistral-7B-Instruct-Embedding-GGUF

sourceHugging Facemitupdated 2y agoView on Hugging Face
15likes3.9kdownloads
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 -->

E5-Mistral-7B-Instruct-Embedding-GGUF

Original Model

intfloat/e5-mistral-7b-instruct

Run with LlamaEdge

  • —LlamaEdge version: v0.8.2 and above
  • —Prompt template
  • —Prompt type: embedding
  • —Context size: 4096
  • —Run as LlamaEdge service
bash
  wasmedge --dir .:. --nn-preload default:GGML:AUTO:e5-mistral-7b-instruct-Q5_K_M.gguf \
    llama-api-server.wasm \
    --prompt-template embedding \
    --ctx-size 4096 \
    --model-name e5-mistral-7b-instruct

Quantized GGUF Models

NameQuant methodBitsSizeUse case
e5-mistral-7b-instruct-Q2_K.ggufQ2_K22.72 GBsmallest, significant quality loss - not recommended for most purposes
e5-mistral-7b-instruct-Q3_K_L.ggufQ3KL33.82 GBsmall, substantial quality loss
e5-mistral-7b-instruct-Q3_K_M.ggufQ3KM33.52 GBvery small, high quality loss
e5-mistral-7b-instruct-Q3_K_S.ggufQ3KS33.16 GBvery small, high quality loss
e5-mistral-7b-instruct-Q4_0.ggufQ4_044.11 GBlegacy; small, very high quality loss - prefer using Q3KM
e5-mistral-7b-instruct-Q4_K_M.ggufQ4KM44.37 GBmedium, balanced quality - recommended
e5-mistral-7b-instruct-Q4_K_S.ggufQ4KS44.14 GBsmall, greater quality loss
e5-mistral-7b-instruct-Q5_0.ggufQ5_055.00 GBlegacy; medium, balanced quality - prefer using Q4KM
e5-mistral-7b-instruct-Q5_K_M.ggufQ5KM55.13 GBlarge, very low quality loss - recommended
e5-mistral-7b-instruct-Q5_K_S.ggufQ5KS55.00 GBlarge, low quality loss - recommended
e5-mistral-7b-instruct-Q6_K.ggufQ6_K65.94 GBvery large, extremely low quality loss
e5-mistral-7b-instruct-Q8_0.ggufQ8_087.7 GBvery large, extremely low quality loss - not recommended
e5-mistral-7b-instruct-f16.gguff16814.5 GBvery large, extremely low quality loss - not recommended

Quantized with llama.cpp b2334