KoboldCpp
A single executable that runs GGUF language models locally, with a web interface and no installation at all.
These buttons open the developer's own site, repository or store listing in a new tab. wares.gg does not host downloads.
1 more ways to get KoboldCpp
Other download sites
Run by third parties, not by KoboldCpp. Prefer the official page above when it works.
About KoboldCpp
KoboldCpp is one self-contained file. You download it, point it at a GGUF model and run it; there is no installer, no Python environment and no dependencies to resolve. It builds on llama.cpp and runs on CPU or GPU with partial or full offload.
What it has grown into is much wider than text: image generation and editing with Stable Diffusion and Flux models, speech to text through Whisper, text to speech, image recognition, and MCP server support with tool calling. It exposes API endpoints compatible with the OpenAI, Ollama, A1111 and ComfyUI conventions, so existing tools can talk to it, and it bundles the KoboldAI Lite interface with memory, world info, characters and several writing modes.
Strengths
- A single executable with no installation and no external dependencies
- Runs on CPU or GPU, with partial offload for machines short on VRAM
- Handles text, image generation, speech to text, text to speech and vision in one binary
- Exposes OpenAI, Ollama, A1111 and ComfyUI compatible API endpoints
Limitations
- You still have to find and download model files yourself
- Performance on CPU alone is slow for larger models
- The bundled interface is dense and aimed at people who already know the terminology
- Official macOS builds are Apple silicon only
Details
- Pricing
- Free · Free and open source under the AGPL.
- License
- AGPL-3.0
- Developer
- KoboldCpp
- Platforms
- Windows, macOS, Linux
- How it runs
- Downloadable app
- Account
- Not required
- Works offline
- Yes
- Best suited for
- Running a local language model on your own hardware with the least possible setup
- Categories
- Local AI tools
- Last verified
- Added
- Provenance
- Selected from the TechWalrus downloads catalog (AI); facts checked against the developer's own pages, 3 sources on file.