Search

DLSS Multi Frame Generation Unlock and controls for any supported MFG title on RTX series GPUs

Xolotl: run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 110 GB RAM): ~33 tok/s on agent tasks, ~41 in fast mode, 512K context. Plus GLM-5.3-Flash, Qwen3.8 and LFM3-Omni. MoE experts tiered across VRAM, pinned RAM and NVMe, with Xolotl Direct NVMe->VRAM. OpenAI + Anthropic API.

GPU tensor unlock source for RTX 4090 and experimental RTX 4070 Ti SUPER; firmware generated locally

Video, Image and GIF upscale/enlarge(Super-Resolution) and Video frame interpolation. Achieved with Waifu2x, Real-ESRGAN, Real-CUGAN, RTX Video Super Resolution VSR, SRMD, RealSR, Anime4K, RIFE, IFRNet, CAIN, DAIN, and ACNet.

Optimized SGLang runtime for Qwen3.8-27B FP8 with DFlash2 and Qwen3.8 Flash-Next NVFP4 with FR-Spec on one NVIDIA RTX PRO 6000 Blackwell 96 GB GPU (SM120): 524K context, HiCache and NIXL.

Pure-Rust inference engine for Qwen GGUF models (Qwen3 to Qwen3.8, dense and MoE up to 770B+). CUDA and Vulkan (NVIDIA, AMD), hybrid CPU+GPU experts, MTP speculative decoding, OpenAI and Anthropic compatible server. Up to ~1.7x llama.cpp decode on a 12 GB RTX 3060.

Vulkan ray tracing glTF 2.0 renderer — RTX path tracer with PBR materials, AI denoising (DLSS, OptiX), and scene editing. NVIDIA open-source sample.

NVIDIA Real-time Denoising (NRD) library

Vanilla Resources Made Available for Minecraft with RTX — Vanilla RTX Development Repository,

Modular extensible low-level render interface (RHI) with higher level extensions

NRD sample and sandbox, demonstrating path tracing and denoising best practices

Streamline Integration Framework

FrankenDriver. Frankenstein Driver. Drivers for video cards RTX 30XXm, RTX 40XXm from aliexpress. Driver for RTX 40XXm, RTX 30XXm, RTX 20XX from aliexpress. Driver for graphics cards with a laptop chip.

MPO Fix for 5000/6000/7000/9000 AMD GPUs & RTX 2000/3000/4000/5000/GTX 1600

Portal 2 - RTX Remix Compatibility Mod

Cross-platform installer for Triton and SageAttention on ComfyUI. Simplifies GPU-accelerated inference setup for Windows users with automated dependency management and RTX 5090 support.

GTAIV RTX Remix Compatibility Mod

Open-source batch OCR workbench — a free, local alternative to ABBYY FineReader. Powered by Ollama + GLM-OCR + PP-DocLayoutV3, ~0.5s/page on RTX 4090. Three-panel editor, layout-aware, PDF/image batch processing, Markdown/Word export. 批量OCR工作台，纯本地运行，免费平替ABBYY，适合书籍文档数字化。

High-efficiency LLM inference engine in C++/CUDA. Run Llama 70B on RTX 3090.

RTX 50-40-30 Series Undervolt Tool

RTX 6000 Pro Wiki — Running Large LLMs (Qwen3.5-397B, Kimi-K2.5, GLM-5) on PCIe GPUs without NVLink

ComfyUI custom nodes for RTX Video Super Resolution, NVIDIA VFX/nvvfx, LTX 2.3, Bernini Prompt Guide, Wan2.2 reference video edit, image/video compare, video preview, loaders, resize, and visual workflow helpers.

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 Gemma 4 26B Gemma 4 31B configs for 1×/2×/4x/8x cards.

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

The definitive vLLM runtime for dual RTX 2080 Ti 22GB + NVLink, delivering Qwen 27B local inference with maximum 200+ tok/s single-request decode with support of FP8 weight ( Join Discord :https://discord.gg/VFqVVySdMS )

One-click DLSS 5 Neural Rendering manager for PC. Swapper, Feeder, OptiScaler, RenoDX, Multipass, ReShade, F8 overlay. Works on RTX 20/30/40/50, AMD RDNA, Intel Arc. DX8-12, Vulkan, OpenGL, emulators.

Run DeepSeek-V4.1-Flash / DeepSeek-V4-Flash and GLM-5.3-Flash on SM89 (Ada / RTX 4090) and SM120 (RTX PRO 6000) with vLLM

TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM

Neural-coprocessor *MGPU Bridge* ReShade add-on that runs DLSS Neural Rendering on a second GPU while the first renders the game. Not SLI — neural post-processing is a terminal stage, so it can be executed on another device entirely. Measured on dual RTX 5060 Ti, with the method and the mistakes published.