legraphista's picture
Upload README.md with huggingface_hub
ac29d2a verified
|
raw
history blame
8.28 kB
metadata
base_model: 01-ai/Yi-Coder-9B
inference: false
library_name: gguf
license: apache-2.0
pipeline_tag: text-generation
quantized_by: legraphista
tags:
  - quantized
  - GGUF
  - quantization
  - imat
  - imatrix
  - static
  - 16bit
  - 8bit
  - 6bit
  - 5bit
  - 4bit
  - 3bit
  - 2bit
  - 1bit

Yi-Coder-9B-IMat-GGUF

Llama.cpp imatrix quantization of 01-ai/Yi-Coder-9B

Original Model: 01-ai/Yi-Coder-9B
Original dtype: BF16 (bfloat16)
Quantized by: llama.cpp b3669
IMatrix dataset: here


Files

IMatrix

Status: βœ… Available
Link: here

Common Quants

Filename Quant type File Size Status Uses IMatrix Is Split
Yi-Coder-9B.Q8_0.gguf Q8_0 9.38GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q6_K.gguf Q6_K 7.25GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q4_K.gguf Q4_K 5.33GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q3_K.gguf Q3_K 4.32GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q2_K.gguf Q2_K 3.35GB βœ… Available 🟒 IMatrix πŸ“¦ No

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
Yi-Coder-9B.BF16.gguf BF16 17.66GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.FP16.gguf F16 17.66GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q8_0.gguf Q8_0 9.38GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q6_K.gguf Q6_K 7.25GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q5_K.gguf Q5_K 6.26GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q5_K_S.gguf Q5_K_S 6.11GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-9B.Q4_K.gguf Q4_K 5.33GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q4_K_S.gguf Q4_K_S 5.07GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ4_NL.gguf IQ4_NL 5.05GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ4_XS.gguf IQ4_XS 4.79GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q3_K.gguf Q3_K 4.32GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q3_K_L.gguf Q3_K_L 4.69GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q3_K_S.gguf Q3_K_S 3.90GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ3_M.gguf IQ3_M 4.06GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ3_S.gguf IQ3_S 3.91GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ3_XS.gguf IQ3_XS 3.72GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ3_XXS.gguf IQ3_XXS 3.47GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q2_K.gguf Q2_K 3.35GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.Q2_K_S.gguf Q2_K_S 3.12GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ2_M.gguf IQ2_M 3.10GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ2_S.gguf IQ2_S 2.88GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ2_XS.gguf IQ2_XS 2.71GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ2_XXS.gguf IQ2_XXS 2.46GB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-9B.IQ1_M IQ1_M - ⏳ Processing 🟒 IMatrix -
Yi-Coder-9B.IQ1_S IQ1_S - ⏳ Processing 🟒 IMatrix -

Downloading using huggingface-cli

If you do not have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Download the specific file you want:

huggingface-cli download legraphista/Yi-Coder-9B-IMat-GGUF --include "Yi-Coder-9B.Q8_0.gguf" --local-dir ./

If the model file is big, it has been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download legraphista/Yi-Coder-9B-IMat-GGUF --include "Yi-Coder-9B.Q8_0/*" --local-dir ./
# see FAQ for merging GGUF's

Inference

Llama.cpp

llama.cpp/main -m Yi-Coder-9B.Q8_0.gguf --color -i -p "prompt here"

FAQ

Why is the IMatrix not applied everywhere?

According to this investigation, it appears that lower quantizations are the only ones that benefit from the imatrix input (as per hellaswag results).

How do I merge a split GGUF?

  1. Make sure you have gguf-split available
  2. Locate your GGUF chunks folder (ex: Yi-Coder-9B.Q8_0)
  3. Run gguf-split --merge Yi-Coder-9B.Q8_0/Yi-Coder-9B.Q8_0-00001-of-XXXXX.gguf Yi-Coder-9B.Q8_0.gguf
    • Make sure to point gguf-split to the first chunk of the split.

Got a suggestion? Ping me @legraphista!