DSH host RSI for local LLMs (currently Qwen3.8 27B): per-use continual self-improvement, counter-reason self-check, critic escalation, and memory retrieval.简单来说根据论文建立的长期反思教训机制，让大模型不再犯错，越用越强！重点是千问3.…

Windows本地bonsai模型部署启动器

Reed 一苇 — 一苇以航，轻渡学海。独立 Windows 学习 Agent，支持角色技能、记忆与 Obsidian 笔记库。

KVMem virtualizes KV memory across inference backends to make ultra-long-context local AI accessible on limited hardware.

Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.

MiniMaxH3_LatentUpscaler for ComfyUI

DNNodes ComfyUI 节点包含「资产卡」等：把包含图/音/描述的角色等资产卡组合，可以配合agent，结合官流或SelfLift二采，在任务台批量运行分段，简洁灵活，或配合搅拌手导演台使用；含提示词富文本编辑器；含SelfLift可一采缓存/二采节点。示例工作流见 example_workflows/

High-performance single-GPU inference for selected model checkpoints and GPUs.

Desktop UI for the NInfer local LLM inference engine

An extension utility for llama.cpp, used with 3090*2 + Strix Halo. llama.cpp的拓展小工具，自用于3090*2 + Strix Halo。

Optimized NINFER runtime and reproducible Qwen3.8-27B true-128K inference for NVIDIA RTX 5080 16GB

A reproducible local LLM benchmarking platform with versioned datasets, deterministic scoring, and auditable reports.

用于启动和管理 llama-server (llama.cpp)的 Tauri 桌面启动器，支持多配置快速切换、粘贴、分享你的启动参数 | A Tauri‑based desktop launcher for starting and managing llama‑server (llama.cpp), featuring quick switching between multiple c…

Video Delta Net hybrid attention for MiniMax-H3 as a native ComfyUI node.

OpenCompass is an LLM evaluation platform, supporting a wide range of models from OpenAI, Anthropic, Gemini, Qwen, GLM, DeepSeek, etc, across 100+ datasets covering knowledge, reasoning, coding, sc…

The easiest way to use MiniMax H3. One compact workflow for T2V, I2V, first/last-frame, and reference video generation, with a unified multi-media input, powerful @ references, and inline dialogue …

High-performance out-of-process NVIDIA DLSS suite (Super Resolution, Frame Generation, Neural Rendering & Depth Anything V2) for ComfyUI

ComfyUI wrapper for VOSR 2.0, a image super resolution model

Neural latent upscaler for Minimax H3 (24ch). Bypasses costly 5B-param VAE decode/encode. Upscale low-res latents directly, then refine. Accelerates high-res video gen, outperforms naive interp.

ComfyTV — the canvas-based app that truly belongs to ComfyUI.

漫剧老李 AIGC 创作全流程 Skill · V7.4.26 多技能套装版（Short-Drama Director Suite）：1 总控 + 12 子技能 · 四层纵深——赛道分流 / 素材采风 / 剧本 / 资产 / 分镜 / 三大打戏引擎 / 投喂 / 质检全链路。六道前置锁 + I 画风锁 · 50 份规则库 · 机检 C1~C33 | H1~H12 + 六道投喂闸门 + P4.…

A comprehensive ComfyUI toolkit for video generation, image editing, and audio-driven lip‑sync, featuring Flux, MiniMax, LTX, Wan2.2 and advanced batch workflows.

Windows Cleaner——专治C盘爆红及各种不服！

🎬 VideoSync - 一键式 AI 视频多语言配音工具。集成 WhisperX (ASR)、Qwen (翻译)、Inde-tts2 (语音克隆) 和 FFmpeg，为 Windows 用户提供开箱即用的本地化视频配音解决方案

Added external text input port that concatenates connected prompt to the main prompt (with blank line separation; falls back to external alone if main prompt is empty).

ComfyUI node to assist with Bernini video editing prompting (Ollama + Prompt Presets)

ComfyUI-PromptRelay with full text output