Yi-Coder-1.5B-IMat-GGUF

Llama.cpp imatrix quantization of 01-ai/Yi-Coder-1.5B

Original Model: 01-ai/Yi-Coder-1.5B
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-1.5B.Q8_0.gguf Q8_0 1.57GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q6_K.gguf Q6_K 1.28GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q4_K.gguf Q4_K 963.67MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q3_K.gguf Q3_K 785.72MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q2_K.gguf Q2_K 634.70MB βœ… Available 🟒 IMatrix πŸ“¦ No

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
Yi-Coder-1.5B.BF16.gguf BF16 2.95GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.FP16.gguf F16 2.95GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q8_0.gguf Q8_0 1.57GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q6_K.gguf Q6_K 1.28GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q5_K.gguf Q5_K 1.10GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q5_K_S.gguf Q5_K_S 1.05GB βœ… Available βšͺ Static πŸ“¦ No
Yi-Coder-1.5B.Q4_K.gguf Q4_K 963.67MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q4_K_S.gguf Q4_K_S 904.18MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ4_NL.gguf IQ4_NL 866.16MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ4_XS.gguf IQ4_XS 832.57MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q3_K.gguf Q3_K 785.72MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q3_K_L.gguf Q3_K_L 826.04MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q3_K_S.gguf Q3_K_S 723.41MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ3_M.gguf IQ3_M 754.38MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ3_S.gguf IQ3_S 723.41MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ3_XS.gguf IQ3_XS 694.95MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ3_XXS.gguf IQ3_XXS 653.90MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q2_K.gguf Q2_K 634.70MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.Q2_K_S.gguf Q2_K_S 618.48MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ2_M.gguf IQ2_M 624.80MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ2_S.gguf IQ2_S 601.60MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ2_XS.gguf IQ2_XS 563.91MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ2_XXS.gguf IQ2_XXS 537.57MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ1_M.gguf IQ1_M 508.57MB βœ… Available 🟒 IMatrix πŸ“¦ No
Yi-Coder-1.5B.IQ1_S.gguf IQ1_S 491.17MB βœ… Available 🟒 IMatrix πŸ“¦ No

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-1.5B-IMat-GGUF --include "Yi-Coder-1.5B.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-1.5B-IMat-GGUF --include "Yi-Coder-1.5B.Q8_0/*" --local-dir ./
# see FAQ for merging GGUF's

Inference

Llama.cpp

llama.cpp/main -m Yi-Coder-1.5B.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-1.5B.Q8_0)
  3. Run gguf-split --merge Yi-Coder-1.5B.Q8_0/Yi-Coder-1.5B.Q8_0-00001-of-XXXXX.gguf Yi-Coder-1.5B.Q8_0.gguf
    • Make sure to point gguf-split to the first chunk of the split.

Got a suggestion? Ping me @legraphista!

Downloads last month
59
GGUF
Model size
1.48B params
Architecture
llama

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Examples
Inference API (serverless) has been turned off for this model.

Model tree for legraphista/Yi-Coder-1.5B-IMat-GGUF

Quantized
(8)
this model