vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-05-09 19:46:11 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

update colab

Browse source
This commit is contained in:
Concedo 2024-06-03 14:55:12 +08:00
parent 8b29d5f848
commit 5ebc532ca9
1 changed files with 11 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

12
colab.ipynb
View file

@ -60,6 +60,10 @@
"LoadImgModel = False #@param {type:\"boolean\"}\r\n",
"ImgModel = \"https://huggingface.co/koboldcpp/imgmodel/resolve/main/imgmodel_ftuned_q4_0.gguf\" #@param [\"https://huggingface.co/koboldcpp/imgmodel/resolve/main/imgmodel_ftuned_q4_0.gguf\"]{allow-input: true}\r\n",
"SCommand = \"\"\r\n",
"#@markdown <hr>\r\n",
"LoadSpeechModel = False #@param {type:\"boolean\"}\r\n",
"SpeechModel = \"https://huggingface.co/koboldcpp/whisper/resolve/main/whisper-base.en-q5_1.bin\" #@param [\"https://huggingface.co/koboldcpp/whisper/resolve/main/whisper-base.en-q5_1.bin\"]{allow-input: true}\r\n",
"WCommand = \"\"\r\n",
"\r\n",
"import os\r\n",
"if not os.path.isfile(\"/opt/bin/nvidia-smi\"):\r\n",
@ -80,6 +84,10 @@
" SCommand = \"--sdmodel imodel.gguf --sdthreads 4 --sdquant --sdclamped\"\r\n",
"else:\r\n",
" SCommand = \"\"\r\n",
"if SpeechModel and LoadSpeechModel:\r\n",
" WCommand = \"--whispermodel wmodel.bin\"\r\n",
"else:\r\n",
" WCommand = \"\"\r\n",
"!echo Finding prebuilt binary for {kvers}\r\n",
"!wget -O dlfile.tmp https://kcppcolab.concedo.workers.dev/?{kvers} && mv dlfile.tmp koboldcpp_cublas.so\r\n",
"!test -f koboldcpp_cublas.so && echo Prebuilt Binary Exists || echo Prebuilt Binary Does Not Exist\r\n",
@ -92,7 +100,9 @@
" !aria2c -x 10 -o vmodel.gguf --summary-interval=5 --download-result=default --allow-overwrite=true --file-allocation=none $LLaVAmmproj\r\n",
"if SCommand:\r\n",
" !aria2c -x 10 -o imodel.gguf --summary-interval=5 --download-result=default --allow-overwrite=true --file-allocation=none $ImgModel\r\n",
"!python koboldcpp.py model.gguf --usecublas 0 mmq --multiuser --gpulayers $Layers --contextsize $ContextSize --quiet --remotetunnel $VCommand $SCommand\r\n"
"if WCommand:\r\n",
" !aria2c -x 10 -o wmodel.bin --summary-interval=5 --download-result=default --allow-overwrite=true --file-allocation=none $SpeechModel\r\n",
"!python koboldcpp.py model.gguf --usecublas 0 mmq --multiuser --gpulayers $Layers --contextsize $ContextSize --quiet --remotetunnel $VCommand $SCommand $WCommand\r\n"
]
}
],