AI image generation

Estúdio de luz para MiniMax H3 dentro do ComfyUI. Lighting studio for MiniMax H3 inside ComfyUI.

One-click blurry/deformed face refinement for MiniMax H3 video / ???? MiniMax H3 ????????

A web tool designed to build Z-Image Turbo prompts and store them in a reusable library.

心软枪法 · Codex Skill — 七宗罪哭戏短视频 Prompt 工厂 | Doubao/Jimeng prompts for Seven Deadly Sins crying-shot short videos

Niyam's Notionaly — monochrome hand-drawn illustration skills for Codex. One ink, tapered linework, transparent background. Generate illustrations, characters, icons and infographics, or redraw your own photo. Portable prompts for ChatGPT, Gemini and Claude included.

dalle-mini (2022) on Apple Silicon via MLX, plus a constrained collage harness for agents to make pictures from its tiles

Interactive trajectory exploration for text2img generation. Intermediate results preview, branch selection, variation controls.

Wes Anderson-inspired poster and cinematic scene skill: formal framing, period art direction, film palettes, scene composition, and title typography.

Curated GPT Image 2.5 (ChatGPT Images 2.5) text-to-image prompts, each with the image it produced — copy & run in Renoise

Content-aware image splitting for multimodal coding agents

GPT Image 2.5 官方提示词指南 中文PPT版本

南鸢·人像角色设定师：把人物气质转成脸部结构、妆容与完整中文提示词

GPT Image 2.5 prompts, verbatim and credited: 150 community prompts across 9 use cases, each with the author who wrote it and their post. Free to run.

MIYANG diagnosis-guided restoration for digital ripple artifacts in iteratively edited AI images

PRINTFILM：AI 科普视频与漫剧创作平台

ローカルMiniMax H3で作るピクセルパーフェクトな動くドット絵 — 日英の手順と検証済み3作例 / Pixel-perfect animated pixel art: bilingual workflow and 3 verified examples

Native NVIDIA DLSS Frame Generation on GeForce RTX 30 in ELDEN RING - DLSS-G on Ampere (sm_86), with no FSR fallback. Combines ERSS-FG with dlssg_for_sm86, adds a settings profile and a verifier that proves from the logs which frame generation actually ran.

A simple paint app with GPT Image 2.5. Draw, select an area, and describe the edit.