A fork of Ultimate SD Upscale Guider with support of MiniMax H3 model
Oh My DSH — tiered model routing for DeepSeek Harness: think/build tiers, automatic vision tier for images, image generation, and an 'oh my dsh' settings tab (web + TUI).
Removing AI watermarks from text, images, and audio/video files.
Maestro AI port to AMD GPU
On-device AI background remover built with React Native, Expo, executorch & Skia.
AI 视觉提示词与案例知识库,收录海报、知识图解、产品概念、插画与城市视觉案例。
High-precision model and CLIP tuning suite for Krea-2 models in ComfyUI
Portable and installable desktop upscaler for Nvidia GPUs supporting models from OpenModelDB
🚧 Under construction — no-training RefMod reference adapters for MiniMax H3 (compress reference images/videos into tiny latents, no training needed).
Agent-native Blender world lock and dual-path image generation for deterministic camera changes
Ultra-clean, minimal Next.js starter that developers can clone and deploy with one click to Vercel or Cloudflare Pages. Get your API token using the link below👇
Sero — on-device AI image detector Chrome extension (ONNX/WASM, MV3)
A Window app that watches your screenshot folder and uses AI to suggest meaningful filenames.
Character turnaround and location reference sheets from a MiniMax-H3 camera move, with vision-model frame selection.
踏雪创意风格:影像指令编译器 skill。7 家族 32 变体风格库,提示词七维优化,元提示词创作,学习记忆。
Local API to use codex programmatically for text and images
inference code for minimax h3