Here is a dlssg for RTX30 Series GPU
THE FIRST DLSS 5 MANAGER
Windows GPU UNLOCK: RTX3090 UEFI port, Korean no-reset Windows boot guide, verified source and benchmark evidence
LLM speculative inference server for heterogeneous hardware & consumer GPUs
支持视频、图片任意输入,借助 DLSS5 算法提升图像与视频渲超分染真实质感;兼容 RTX30/40/50 系列 NVIDIA 显卡,支持 HDR、队列批量处理,提供实时渲染预览与参数调节。可保留原始格式、分辨率及容器编码,也可自定义无损输出;Windows 端开箱即用,无需游戏引擎,不依赖额外组件。
Qwen3.8-Flash-Next on 2× RTX 3090: with 128 GB RAM up to 4,191 tok/s prefill · 111.5 tok/s decode (131K prompt), full 256K window at 2,865 tok/s; with 64 GB RAM 3,410 tok/s prefill · 84 tok/s decode. New: opt-in uncensored mode (runtime abliteration, no new weights).
RTX 30 系 DLSS 帧生成管理器:按游戏部署 dlssg_for_sm86,一键恢复,自动检测内核级反作弊 | DLSS frame generation manager for RTX 30 series: per-game deployment of dlssg_for_sm86, one-click restore, kernel-level anti-cheat detection
RTX 3060 12GB/8GB 本地跑 27B 大模型:256K 上下文、约 45 token/秒,能看图、调工具,附懒人包和源码
Enable NVIDIA Smooth Motion on RTX 30-series / Ampere SM86 - one-click Windows utility, no per-game DLSS-FG integration required.
Unlocking NVIDIA Smooth Motion (DLSS Frame Generation) on Ampere (RTX 30 Series / SM86) via In-Memory Re-Hosting & Driver Interception
RouteWeaver: experimental cache-safe speculative inference and CPU/GPU transfer optimization for Qwen 27B on low-VRAM NVIDIA GPUs.
Qwen3.8-27B EXL3 (4.00 bpw) + DFlash2 speculative decoding for ExLlamaV3, validated at 262k context on a 24 GB RTX 3090