Open Issues Need Help
View All on GitHub Flatpak Support 23 days ago
help wanted
LostRuins/koboldcpp
11.6K
Run GGUF models easily with a KoboldAI UI. One File. Zero Install.
C++
#gemma#ggml#gguf#koboldai#koboldcpp#language-model#llama#llamacpp#llm#mistral
Windows Vulkan MTP is 44% slower than llama.cpp 3 months ago
help wanted
LostRuins/koboldcpp
11.6K
Run GGUF models easily with a KoboldAI UI. One File. Zero Install.
C++
#gemma#ggml#gguf#koboldai#koboldcpp#language-model#llama#llamacpp#llm#mistral
After launcher window resized in Linux, tooltips popup windows become so narrow they cannot show a single letter 6 months ago
bug help wanted
LostRuins/koboldcpp
11.6K
Run GGUF models easily with a KoboldAI UI. One File. Zero Install.
C++
#gemma#ggml#gguf#koboldai#koboldcpp#language-model#llama#llamacpp#llm#mistral
Notable performance regression in token generation speed (vulkan, windows FlashAttention). 8 months ago
help wanted
LostRuins/koboldcpp
11.6K
Run GGUF models easily with a KoboldAI UI. One File. Zero Install.
C++
#gemma#ggml#gguf#koboldai#koboldcpp#language-model#llama#llamacpp#llm#mistral
GLM 4.0 Was Broken in KoboldCPP 1.100.X, Possible Degradation Also From Prior Versions 10 months ago
help wanted
LostRuins/koboldcpp
11.6K
Run GGUF models easily with a KoboldAI UI. One File. Zero Install.
C++
#gemma#ggml#gguf#koboldai#koboldcpp#language-model#llama#llamacpp#llm#mistral