llama.cpp
LLM inference engine in C/C++ for running large language models locally.
0 members list llama.cpp, unchanged over the last 12 weeks, data as of 12 November 2026
Members who list llama.cpp in their Stack, by week.
Members who list llama.cpp
Be the first to add llama.cpp
Nobody on Stackness runs llama.cpp yet. Add it to your stack and you're the first member running it.
Homebrew analytics installs: 31,970 installs (2026) to 35,810 installs (2026)
Imported from Homebrew analytics, last reading Oct 2026
Wikipedia pageviews: 2,256 views (2024) to 13,691 views (2026)
Imported from Wikipedia pageviews, last reading Oct 2026
About llama.cpp
llama.cpp is an inference engine for running large language models locally on your computer using C and C++. The tool enables developers to execute frontier AI models on their own machines without relying on API keys, telemetry, or cloud services, keeping files and requests entirely local. It is optimized to work across diverse hardware configurations, from laptops to clusters, supporting various GPUs and CPUs with the same binary and hand-tuned kernels.
The tool is developed by ggml-org and released under the MIT License. It is available on FreeBSD, Linux, Microsoft Windows, and macOS, with Python bindings also provided. The project offers straightforward setup through package managers or source compilation, allowing developers to serve models and integrate them with local coding agents without requiring configuration or API credentials.
Checked 8 November 2026.
Nobody on Stackness runs llama.cpp yet.
Sign up and add itPosts about llama.cpp
- Claude Code alternatives: add Codex first, and the three agents that replace it
Three agents replace Claude Code as of November 2026, but 5 of 8 Stackness members who use it run a second agent beside it. Add Codex before you leave.
- Strands Decider 2B: what it is and where it sits in an agent harness
Strands Decider 2B is AWS's free 1.9B decision model, out 1 October 2026: about 115 ms per question on an RTX 3090, consulted before each tool call.
- Mistral Large 4 vs Beam vs Kolibri: which open weights you can download, and what your hardware can serve
Only Kolibri's weights are out. Beam and Mistral Large 4 stay API previews until late October, and Kolibri's 4-bit build runs at 50 tok/s on a 64GB Mac.