Fused TBQ4 Flash Attention + MTP + Shared Tensors for llama.cpp — 82+ tok/s with lossless 4.25 bpv KV cache at 200K context on RTX 4090
GitHub repository with 78 stars and 5 forks.
Language: C++
Topics: cuda, flash-attention, fwht, kv-cache, llama-cpp, mtp, multi-token-prediction, quantization, qwen, rtx-4090
Trending score 0.08, activity score 0.00, stars gained +0, forks gained +0.
2026-06-04: 78 stars and 5 forks.
:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.
GitHub repository with 1,493 stars and 480 forks.
Trending score: 1.66; stars gained: +5; forks gained: +3.
Language: C++
Topics: accelerator, ai, cuda, deepseek, gpu, img-gen
Aim-bot based on AI for all FPS and TPS games
GitHub repository with 514 stars and 105 forks.
Trending score: 1.48; stars gained: +4; forks gained: +1.
Language: C++
Topics: ai, ai-aimbot, aimbot, arduino, cpp, cs2
Fast LLM speculative inference server for consumer hardware.
GitHub repository with 2,329 stars and 217 forks.
Trending score: 1.08; stars gained: +9; forks gained: +1.
Language: C++
Topics: cuda, cuda-kernels, dflash, kernel, llama-cpp, local-ai
Making it easier to work with shaders
GitHub repository with 5,348 stars and 451 forks.
Trending score: 0.76; stars gained: +5; forks gained: +1.
Language: C++
Topics: cuda, d3d12, glsl, hlsl, shaders, vulkan
Train, inspect, edit, automate, and export 3D Gaussian Splatting scenes from a single native application.
GitHub repository with 3,161 stars and 344 forks.
Trending score: 0.76; stars gained: +5; forks gained: +0.
Language: C++
Topics: computer-graphics, computer-vision, cuda, gaussian-splatting, optimization
A retargetable MLIR-based machine learning compiler and runtime toolkit.
GitHub repository with 3,787 stars and 918 forks.
Trending score: 0.74; stars gained: +3; forks gained: +0.
Language: C++
Topics: mlir, vulkan, tensorflow, spirv, cuda, jax
LLM inference in C/C++
GitHub repository with 114,606 stars and 19,170 forks.
Trending score: 4.00; stars gained: +164; forks gained: +36.
Language: C++
Topics: ggml
DuckDB is an analytical in-process SQL database management system
GitHub repository with 38,604 stars and 3,293 forks.
Trending score: 2.81; stars gained: +24; forks gained: +2.
Language: C++
Topics: analytics, database, embedded-database, olap, sql
GitHub repository with 6,791 stars and 5,142 forks.
Trending score: 2.73; stars gained: +25; forks gained: +9.
Language: C++
ClickHouse® is a real-time analytics database management system
GitHub repository with 47,812 stars and 8,466 forks.
Trending score: 2.61; stars gained: +24; forks gained: +4.
Language: C++
Topics: ai, analytics, big-data, clickhouse, cloud-native, cpp
An Open Source Machine Learning Framework for Everyone
GitHub repository with 195,410 stars and 75,345 forks.
Trending score: 2.39; stars gained: +31; forks gained: +1.
Language: C++
Topics: deep-learning, deep-neural-networks, distributed, machine-learning, ml, neural-network
Godot Engine – Multi-platform 2D and 3D game engine
GitHub repository with 112,059 stars and 25,554 forks.
Trending score: 2.37; stars gained: +332; forks gained: +24.
Language: C++
Topics: game-engine, godot, godotengine, open-source, multi-platform, gamedev
SGLang is a high-performance serving framework for large language models and multimodal models.
GitHub repository with 28,872 stars and 6,332 forks.
Trending score: 2.89; stars gained: +14; forks gained: +31.
Language: Python
Topics: attention, blackwell, cuda, deepseek, diffusion, glm
Use your NVIDIA GPU's VRAM as swap space on Linux. Built for laptops with soldered memory and no upgrade path. If you have an RTX card sitting there with 8GB of VRAM and you're getting swapped to SSD, this puts that VRAM to work
GitHub repository with 385 stars and 10 forks.
Trending score: 2.51; stars gained: +220; forks gained: +5.
Language: Shell
Topics: cuda, gpu, laptop, linux, memory, nbd
A high-throughput and memory-efficient inference and serving engine for LLMs
GitHub repository with 81,936 stars and 17,651 forks.
Trending score: 2.08; stars gained: +131; forks gained: +40.
Language: Python
Topics: amd, blackwell, cuda, deepseek, deepseek-v3, gpt
Pure Rust Inference Engine
GitHub repository with 472 stars and 65 forks.
Trending score: 1.99; stars gained: +10; forks gained: +1.
Language: Rust
Topics: cuda, dgx, dgx-spark, gb10, llm-inference, mamba
A GPU cluster manager that configures and orchestrates inference engines like vLLM and SGLang for high-performance AI model deployment.
GitHub repository with 5,100 stars and 541 forks.
Trending score: 1.85; stars gained: +10; forks gained: +2.
Language: Python
Topics: ascend, cuda, deepseek, distributed-inference, genai, inference
:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.
GitHub repository with 1,493 stars and 480 forks.
Trending score: 1.66; stars gained: +5; forks gained: +3.
Language: C++
Topics: accelerator, ai, cuda, deepseek, gpu, img-gen