Popular repositories Loading

1. dot dot Public

   Forked from haras-unicorn/dot

   Configurations for my NixOS systems.

   Nix

2. llama-moe-cache llama-moe-cache Public

   Forked from ongunm/llama-moe-cache

   Expert cache + predictive prefetch for MoE inference in llama.cpp. A 12GB GPU can run a 120GB model at native speed. At 3% sparsity, a single workstation could theoretically run a 1T parameter mode…

   C++

3. llama.cpp llama.cpp Public

   Forked from haras-unicorn/llama.cpp

   LLM inference in C/C++

   C++

4. mcp-rss mcp-rss Public

   Forked from haras-unicorn/mcp-rss

   MCP server that provides RSS tooling

   Nix

5. mcp-plan mcp-plan Public

   Forked from haras-unicorn/mcp-plan

   MCP server that provides planning tooling.

   Nix