LostRuins
individualScores
1 product on the map — 1 open.
Openness
5 high confidence- license
- AGPL-3.0 (OSI, KoboldCpp and the bundled KoboldAI Lite UI; bundled GGML, llama.cpp and other components keep their MIT or Apache-2.0 licenses)
- source
- public (the release executables build from the repository)
- core features withheld
- no — an individual developer's project with no company or paid version behind it
KoboldCpp and its bundled web interface are AGPL-3.0, and the components it builds on, llama.cpp among them, keep their own permissive licenses. It is one developer's project with no company or paid version behind it, and the release executables build from the public repository.
- https://raw.githubusercontent.com/LostRuins/koboldcpp/concedo/LICENSE.md recorded 2026-11-12
The LICENSE.md body is the GNU Affero General Public License version 3 text.
- https://raw.githubusercontent.com/LostRuins/koboldcpp/HEAD/README.md recorded 2026-11-12
The README says "KoboldCpp and KoboldAI Lite are licensed under the GNU AGPL v3.0, unless a file states otherwise. Bundled components retain their respective licenses, including the MIT license for GGML, llama.cpp", calls the project "free and open-source software", and documents building from source. Searched for pricing, paid and license key: none.
- https://ungh.cc/repos/LostRuins/koboldcpp/files/concedo recorded 2026-11-12
The repository file listing shows koboldcpp.py, the C++ sources, the build scripts and per-component license files, and no ee, enterprise, commercial or proprietary path.
Adoption
3 low confidenceKoboldCpp ships as standalone executables on its GitHub releases page rather than as a package, so there is no download count to measure. GitHub stars are the only signal, and a star is not a use.
- https://ungh.cc/repos/LostRuins/koboldcpp recorded 2026-11-12
"stars":12209,"forks":807, pushed 2026-11-09, for LostRuins/koboldcpp
Capability
3 medium confidenceKoboldCpp packs local text, image, speech and music generation, an agent and several compatible APIs into one executable for a single machine. It runs llama.cpp underneath and offers none of the multi-GPU or datacenter serving vLLM is built for, so it sits two steps below vLLM.
- https://raw.githubusercontent.com/LostRuins/koboldcpp/HEAD/README.md recorded 2026-11-12
The README's feature list covers "Local text generation" of GGUF models, image and video generation, speech-to-text with Whisper, text to speech, music generation, "MCP server support, tool calling, web search, retrieval-augmented generation", endpoints for "KoboldAI, OpenAI, Ollama, A1111/Forge, ComfyUI", and a "Single standalone executable for Windows, Linux or macOS".