Stars
Give GeForce cards (Turing and later) a BAR1 at least as large as their VRAM on Linux, without flashing.
Screen, input, audio, microphone, USB and clipboard between Linux machines over RDMA (RoCE): gamescope streaming with VRR, KMS scanout, multiseat in LXC
MCP server for searching secondhand marketplaces — Facebook Marketplace, eBay, and Depop. Works with Claude Desktop, Claude Code, and other MCP clients.
CarPlay Ultra & Android Auto head unit for Linux, macOS and Android, built in Rust + Slint. Multi-session wired + wireless, zero-copy hardware video, low-latency audio.
efschu / llama-swap
Forked from mostlygeek/llama-swapReliable model swapping for any local OpenAI/Anthropic compatible server - llama.cpp, vllm, etc
efschu / vs-rife
Forked from HolyWu/vs-rifeRIFE function for VapourSynth
efschu / proxycache
Forked from dingausmwald/proxycacheSmart OpenAI‑compatible proxy for llama.cpp: manages slots, saves/restores KV cache to disk, routes requests by prefix similarity, and protects hot slots from being overwritten. Accelerates long pr…
efschu / oh-my-pi
Forked from can1357/oh-my-pi⌥ AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more
Diffusion model(SD,Flux,Wan,Qwen Image,Z-Image,...) inference in pure C/C++
Private fork of fewtarius/CachyLLama - CUDA FlashAttention KV-quant fix + 2x Tesla V100 (sm_70) build and deployment notes
Vendor-neutral collective communication for GPU groups without a common native collective library, plus a PCIe/BAR1 direct path for consumer cards where peer-to-peer is locked out
[CVPR 2026] Towards Real-Time Diffusion-Based Streaming Video Super-Resolution — An efficient one-step diffusion framework for streaming VSR with locality-constrained sparse attention and a tiny co…
efschu / htsglang
Forked from sgl-project/sglanghtsglang: heterogeneous, tier-aware sglang — uneven compute + VRAM pooling, KV spill to RAM, adaptive drafter routing, improved GGUF and more
efschu / shvllm
Forked from vllm-project/vllmA high-throughput and memory-efficient inference and serving engine for LLMs
vLLM Quantization plugin for GGUF
KVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. Calibration-free, one flag.
Community recipes for serving LLMs on RTX 3090/4090/5090 CUDA gpus. Multi-engine (vLLM, sglang, llama.cpp and other custom engines) and model-agnostic. Currently shipping Qwen3.8-27B Qwen3.6 35B Ge…
efschu / vsrawsource
Forked from chikuzen/vsrawsourceRaw format video reader for VapourSynth
efschu / FFmpeg
Forked from FFmpeg/FFmpegMirror of https://git.ffmpeg.org/ffmpeg.git
efschu / KVarN
Forked from huawei-csl/KVarNKVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. Calibration-free, one flag.
VapourSynth TensorRT plugin with multi-engine pipeline, RTX support, and Lanczos-3 resize
efschu / jellyfin
Forked from jellyfin/jellyfinThe Free Software Media System - Server Backend & API