๐ Model Description
language:
- ja
google/gemma-3-12b-it-qat-q40-unquantizedใๆฅๆฌ่ชใๅคใๅซใพใใimatrixใไฝฟใฃใฆ้ๅญๅใใใขใใซใงใ
This is a model that quantizes google/gemma-3-12b-it-qat-q40-unquantized using an imatrix that contains a lot of Japanese..
https://huggingface.co/dahara1/imatrix-jpn-test).
ๆๆฐใฎllama.cppใไฝฟใฃใฆๅใใใฆใใ ใใใ
Please use the latest llama.cpp.
llama-mtmd-cliใณใใณใใจmmproj.ggufใใกใคใซใไฝฟใใจ็ปๅใ่ชญใฟใใไบใใงใใพใ
You can use llama-mtmd-cli for image reading.
llama-mtmd-cli -m gemma-3-4b-it-qat-q40-japanese-imatrix-Q4K_L.gguf --mmproj mmproj.gguf --image ./test.png -p "ใใฎ็ปๅใฏใชใใงใใ๏ผ(What is this image?)"
auto download command example.
llama-server.exe -hf dahara1/gemma-3-12b-it-qat-japanese-imatrix:gemma-3-12b-it-qat-q40-japanese-imatrix-Q40.gguf
then access to http://127.0.0.1:8080
๐ GGUF File List
| ๐ Filename | ๐ฆ Size | โก Download |
|---|---|---|
|
gemma-3-12B-it-qat-unquantized-BF16.gguf
LFS
FP16
|
21.92 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ2_M.gguf
Recommended
LFS
Q2
|
4.01 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ2_XS.gguf
LFS
Q2
|
3.58 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ2_XXS.gguf
LFS
Q2
|
3.28 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ3_M.gguf
LFS
Q3
|
5.27 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ3_XS.gguf
LFS
Q3
|
4.85 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ3_XXS.gguf
LFS
Q3
|
4.46 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-IQ4_XS.gguf
LFS
Q4
|
6.1 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q4_0.gguf
LFS
Q4
|
6.41 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q4_K-f16.gguf
LFS
Q4
|
7.91 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q4_K_L.gguf
LFS
Q4
|
7.03 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q4_K_M.gguf
LFS
Q4
|
6.8 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q4_K_S.gguf
LFS
Q4
|
6.46 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q5_K-f16.gguf
LFS
Q4
|
8.97 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q5_K_L.gguf
LFS
Q4
|
8.09 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q5_K_M.gguf
LFS
Q4
|
7.87 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q5_K_S.gguf
LFS
Q4
|
7.67 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q6_K-f16.gguf
LFS
Q4
|
10.1 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q6_K.gguf
LFS
Q4
|
9 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q6_K_L.gguf
LFS
Q4
|
9.22 GB | Download |
|
gemma-3-12b-it-qat-q4_0-japanese-imatrix-Q8_0.gguf
LFS
Q4
|
12.53 GB | Download |
|
mmproj.gguf
LFS
|
814.63 MB | Download |