Edit model card

Gemma-2-9B-It-SPPO-Iter3-IMat-GGUF

Llama.cpp imatrix quantization of UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3

Original Model: UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
Original dtype: BF16 (bfloat16)
Quantized by: llama.cpp b3311
IMatrix dataset: here


Files

IMatrix

Status: βœ… Available
Link: here

Common Quants

Filename Quant type File Size Status Uses IMatrix Is Split
Gemma-2-9B-It-SPPO-Iter3.Q8_0.gguf Q8_0 9.83GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q6_K.gguf Q6_K 7.59GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q4_K.gguf Q4_K 5.76GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q3_K.gguf Q3_K 4.76GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q2_K.gguf Q2_K 3.81GB βœ… Available 🟒 IMatrix πŸ“¦ No

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
Gemma-2-9B-It-SPPO-Iter3.BF16.gguf BF16 18.49GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.FP16.gguf F16 18.49GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q8_0.gguf Q8_0 9.83GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q6_K.gguf Q6_K 7.59GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q5_K.gguf Q5_K 6.65GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q5_K_S.gguf Q5_K_S 6.48GB βœ… Available βšͺ Static πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q4_K.gguf Q4_K 5.76GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q4_K_S.gguf Q4_K_S 5.48GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ4_NL.gguf IQ4_NL 5.44GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ4_XS.gguf IQ4_XS 5.18GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q3_K.gguf Q3_K 4.76GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q3_K_L.gguf Q3_K_L 5.13GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q3_K_S.gguf Q3_K_S 4.34GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ3_M.gguf IQ3_M 4.49GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ3_S.gguf IQ3_S 4.34GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ3_XS.gguf IQ3_XS 4.14GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ3_XXS.gguf IQ3_XXS 3.80GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q2_K.gguf Q2_K 3.81GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.Q2_K_S.gguf Q2_K_S 3.55GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ2_M.gguf IQ2_M 3.43GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ2_S.gguf IQ2_S 3.21GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ2_XS.gguf IQ2_XS 3.07GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ2_XXS.gguf IQ2_XXS 2.82GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ1_M.gguf IQ1_M 2.55GB βœ… Available 🟒 IMatrix πŸ“¦ No
Gemma-2-9B-It-SPPO-Iter3.IQ1_S.gguf IQ1_S 2.38GB 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/Gemma-2-9B-It-SPPO-Iter3-IMat-GGUF --include "Gemma-2-9B-It-SPPO-Iter3.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/Gemma-2-9B-It-SPPO-Iter3-IMat-GGUF --include "Gemma-2-9B-It-SPPO-Iter3.Q8_0/*" --local-dir ./
# see FAQ for merging GGUF's

Inference

Simple chat template

<bos><start_of_turn>user
{user_prompt}<end_of_turn>
<start_of_turn>model
{assistant_response}<end_of_turn>
<start_of_turn>user
{next_user_prompt}<end_of_turn>

Llama.cpp

llama.cpp/main -m Gemma-2-9B-It-SPPO-Iter3.Q8_0.gguf --color -i -p "prompt here (according to the chat template)"

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: Gemma-2-9B-It-SPPO-Iter3.Q8_0)
  3. Run gguf-split --merge Gemma-2-9B-It-SPPO-Iter3.Q8_0/Gemma-2-9B-It-SPPO-Iter3.Q8_0-00001-of-XXXXX.gguf Gemma-2-9B-It-SPPO-Iter3.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
51,604
GGUF
Model size
9.24B params
Architecture
gemma2

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

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

Quantized from

Dataset used to train legraphista/Gemma-2-9B-It-SPPO-Iter3-IMat-GGUF