RichardErkhov's picture
uploaded readme
d381bf0 verified

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SoRu-0003 - GGUF

Name Quant method Size
SoRu-0003.Q2_K.gguf Q2_K 0.32GB
SoRu-0003.IQ3_XS.gguf IQ3_XS 0.32GB
SoRu-0003.IQ3_S.gguf IQ3_S 0.32GB
SoRu-0003.Q3_K_S.gguf Q3_K_S 0.31GB
SoRu-0003.IQ3_M.gguf IQ3_M 0.32GB
SoRu-0003.Q3_K.gguf Q3_K 0.33GB
SoRu-0003.Q3_K_M.gguf Q3_K_M 0.33GB
SoRu-0003.Q3_K_L.gguf Q3_K_L 0.34GB
SoRu-0003.IQ4_XS.gguf IQ4_XS 0.33GB
SoRu-0003.Q4_0.gguf Q4_0 0.33GB
SoRu-0003.IQ4_NL.gguf IQ4_NL 0.33GB
SoRu-0003.Q4_K_S.gguf Q4_K_S 0.36GB
SoRu-0003.Q4_K.gguf Q4_K 0.37GB
SoRu-0003.Q4_K_M.gguf Q4_K_M 0.37GB
SoRu-0003.Q4_1.gguf Q4_1 0.35GB
SoRu-0003.Q5_0.gguf Q5_0 0.37GB
SoRu-0003.Q5_K_S.gguf Q5_K_S 0.38GB
SoRu-0003.Q5_K.gguf Q5_K 0.39GB
SoRu-0003.Q5_K_M.gguf Q5_K_M 0.39GB
SoRu-0003.Q5_1.gguf Q5_1 0.39GB
SoRu-0003.Q6_K.gguf Q6_K 0.47GB
SoRu-0003.Q8_0.gguf Q8_0 0.49GB

Original model description:

base_model: ehristoforu/SoRu-0002 tags: - text-generation-inference - transformers - unsloth - qwen2 - trl - sft license: apache-2.0 language: - en

Uploaded model

  • Developed by: ehristoforu
  • License: apache-2.0
  • Finetuned from model : ehristoforu/SoRu-0002

This qwen2 model was trained 2x faster with Unsloth and Huggingface's TRL library.