KoboldCpp
Easy-to-use AI text generation with llama.cpp backend
F
Score: 38/100
Type
Execution
hybrid
Interface
gui
About
KoboldCpp is a simple one-file way to run various GGML and GGUF models. It combines a llama.cpp backend with an easy-to-use API and web interface. Features include automatic GPU layer detection, context shifting, and story/chat modes.
Performance
1500ms
Cold Start
400MB
Base Memory
300ms
Startup Overhead
✓ Last Verified
Date: Jan 18, 2026
Method: manual test
Manually verified
Languages
C++Python
Details
- Isolation
- process
- Maturity
- stable
- License
- AGPL-3.0