NInfer for Windows and around 16GB VRAM: RTX 5070 Ti / 5080 / 5090, Qwen3.8-27B GSQ-RCO Q3, CUDA 13 Native engine, tray manager, model conversion and measured setup guides.
Crazy Fast Qwen3.8–27b With a 420k Context on an RTX 5090
Local, portable GUI for Qwen3-TTS. Optimized for NVIDIA RTX 50 Series (CUDA 12.8). One-click install.
A hands-on guide for AI builders: make your own RTX 4090D/5090 GPU server that’s fast and efficient.
Qwen3.8-Flash-Next 177B MoE on RTX 5090 Laptop (24GB VRAM + 64GB RAM): llama.cpp 22-25 tok/s -> Strata 93.5 tok/s (3.7x), GPU+CPU both saturated. 7 quant tiers screened over 26 rounds, MTP corruption found & fixed, 256K context + vision. 中英双语实测实录
A local Gradio Web-UI for the VIBE model (https://huggingface.co/iitolstykh/VIBE-Image-Edit), supporting advanced features...
NVFP4 LLM inference tuned for consumer Blackwell GPUs. Auto-detects your GPU, downloads the model, serves an OpenAI-compatible API — benchmarked and quality-certified.
Optimized vLLM deployment for NVIDIA Blackwell (RTX 5090) on Linux Kernel 6.14. Resolves SM_120 kernel incompatibilities, P2P deadlocks, and memory fragmentation for high-performance LLM inference.
A lightweight, high‑performance local interface for **FLUX.2 Klein (4B)** by Black Forest Labs.
⚡ A lightning-fast local Gradio GUI for Z-Image-Turbo (Windows). Optimized for NVIDIA RTX 5090 (Blackwell) & CUDA 12.8.
AI-generated RTX 5090 XBAR overclocking tool with response code for related branches/discussions. Tested only on the author's configuration. For other setups, follow the docs and use an AI agent to adapt it to your machine. Use at your own risk.
A lightweight, high-performance local interface for **FLUX.2 Klein (9B)** by Black Forest Labs. This project is specifically optimized for **NVIDIA RTX 5090** GPUs, utilizing PyTorch Nightly (CUDA 13.0) and `bfloat16` precision for maximum speed.
Environment setup and build fixes for 3D Gaussian Splatting on RTX 5090 (Windows, CUDA 12.8) ; See setup instructions in README_rtx5090setup.md
Home AI inference lab running Qwen3.8 across RTX 5090, RTX 3090, and DGX Spark with vLLM, llama.cpp, custom Go routing, benchmarking, monitoring, long-context inference, and Claude Code integration.
Field notes: a Claude Code + Pinokio session installing a full 520+ app cinema/AI workstation on Windows RTX 5090, zero manual clicks
Enable utility for MSI Display Device when the onboard display turns black because the Windows device is disabled.
To associate your repository with the rtx5090 topic, visit your repo's landing page and select "manage topics."