CoolFace
Modelpublic

RichardErkhov/objects76_-_ft-Llama-3.2-3B-Instruct-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes802downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

ft-Llama-3.2-3B-Instruct - GGUF

  • —Model creator: https://huggingface.co/objects76/
  • —Original model: https://huggingface.co/objects76/ft-Llama-3.2-3B-Instruct/
NameQuant methodSize
ft-Llama-3.2-3B-Instruct.Q2_K.ggufQ2_K1.27GB
ft-Llama-3.2-3B-Instruct.IQ3_XS.ggufIQ3_XS1.38GB
ft-Llama-3.2-3B-Instruct.IQ3_S.ggufIQ3_S1.44GB
ft-Llama-3.2-3B-Instruct.Q3_K_S.ggufQ3KS1.44GB
ft-Llama-3.2-3B-Instruct.IQ3_M.ggufIQ3_M1.49GB
ft-Llama-3.2-3B-Instruct.Q3_K.ggufQ3_K1.57GB
ft-Llama-3.2-3B-Instruct.Q3_K_M.ggufQ3KM1.57GB
ft-Llama-3.2-3B-Instruct.Q3_K_L.ggufQ3KL1.69GB
ft-Llama-3.2-3B-Instruct.IQ4_XS.ggufIQ4_XS1.71GB
ft-Llama-3.2-3B-Instruct.Q4_0.ggufQ4_01.79GB
ft-Llama-3.2-3B-Instruct.IQ4_NL.ggufIQ4_NL1.79GB
ft-Llama-3.2-3B-Instruct.Q4_K_S.ggufQ4KS1.8GB
ft-Llama-3.2-3B-Instruct.Q4_K.ggufQ4_K1.88GB
ft-Llama-3.2-3B-Instruct.Q4_K_M.ggufQ4KM1.88GB
ft-Llama-3.2-3B-Instruct.Q4_1.ggufQ4_11.95GB
ft-Llama-3.2-3B-Instruct.Q5_0.ggufQ5_02.11GB
ft-Llama-3.2-3B-Instruct.Q5_K_S.ggufQ5KS2.11GB
ft-Llama-3.2-3B-Instruct.Q5_K.ggufQ5_K2.16GB
ft-Llama-3.2-3B-Instruct.Q5_K_M.ggufQ5KM2.16GB
ft-Llama-3.2-3B-Instruct.Q5_1.ggufQ5_12.28GB
ft-Llama-3.2-3B-Instruct.Q6_K.ggufQ6_K2.46GB
ft-Llama-3.2-3B-Instruct.Q8_0.ggufQ8_03.19GB

Original model description: --- base_model: unsloth/Llama-3.2-3B-Instruct language:

  • —en license: apache-2.0 tags:
  • —text-generation-inference
  • —transformers
  • —unsloth
  • —llama
  • —trl
  • —sft ---

Uploaded model

  • —Developed by: objects76
  • —License: apache-2.0
  • —Finetuned from model : unsloth/Llama-3.2-3B-Instruct

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>

Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more quants, at much higher speed, than I would otherwise be able to.