Just added support for FP8-FULL (previously only FP8-Dev was supported!) Commit: https://github.com/cocktailpeanut/hidream-o1/commit/ce9b...
HiDream O1 Full FP8 Support
Fully Local Video Generation Agent comes to WanGP
See more here: https://x.com/cocktailpeanut/status/2036509028088512533?s=20 WanGP now ships with a native LOCAL LLM agent. Yes, it's FULL...
Run HiDream-o1-image with Low VRAM (10GB)
This launcher lets you run the official web ui from the HiDream-o1-image project at https://github.com/HiDream-ai/HiDream-O1-Image with l...
HiDream-O1-Image: a single transformer image generation model with reasoning skills
What is architecturally novel? HiDream-O1-Image is a single transformer that works directly on raw pixels. - no VAE - no separate text en...
ds4.c reasoning in action -- reasons better than me!
I tried some interesting questions to test the reasoning capabilities of deepseek4 (powered by ds4 ). Here's one image that sums it up: A...
ds4.c reasoning in action -- reasons better than me!
I tried some interesting questions to test the reasoning capabilities of deepseek4 (powered by ds4 ). Here's one image that sums it up: And here's the realtime reasoning process...
ds4-webui: DeepSeek v4 Web UI for Apple Devices with 128GB+ memory
ds4-webui is a web ui for running DeepSeek V4 Flash locally through ds4.c From a user's point of view, it gives you a one-click local wor...
Overworld Low VRAM Word Model Simulator
Belated post on Overworld launcher--This app runs the Overworld Biome server, on NVIDIA GPU machines with at 8GB VRAM. Note that the init...
[Research] ds4 Models and System Requirements
ds4.c is a narrow Metal inference engine for DeepSeek V4 Flash. It is not a general GGUF runner and only supports the GGUF files publishe...
Retrosite: website history visualizer
Retrosite is a local app for making visual timelines from old website captures. Type in a domain or page, and it helps turn the site's Wa...
Reclip supported services
Was curious which sites are exactly supported. Turns out a lot. ReClip uses yt-dlp under the hood — one of the most comprehensive open-so...
Phosphene image2video + text2video examples
Text-to-video Took around 16 minutes on my Mac M1 Max 64GB, but I'm sure newer machines are faster! Image to video Prompt: "He angrily wa...
Phosphene — local video and audio generation for Apple Silicon
Phosphene is a free desktop panel for generating video on Apple Silicon Macs. It wraps Lightricks' LTX 2.3 model running natively on Appl...
Texturizer lets you generate custom textured rigged mesh from any image and a seed mesh
Texturizer is a gradio app that lets you: 1. Take any rigged mesh 2. Add any custom image 3. To generate a textured rigged mesh Needs around 8GB VRAM NVIDIA GPU, low VRAM powere...
Texturizer lets you generate custom textured rigged mesh from any image and a seed mesh
Texturizer is a gradio app that lets you: 1. Take any rigged mesh 2. Add any custom image 3. To generate a textured rigged mesh Needs aro...
AI4AnimationPy: Every frame generated, not authored.
I stumbled on this repo a few days ago and genuinely couldn't believe it wasn't all over my feed. So I built a way for you to try it your...
Stripper: strip out "Made with AI" tags from images
I made a web app + agent-ready CLI for stripping out the "Made with AI" tag from images you upload to X (along with all other user-identi...
Euphony: browse all your Codex chat sessions in a web UI.
OpenAI just open sourced Euphony, a minimal web UI to browse and search all your Codex sessions . Works on ALL OS (Mac, Windows, Linux) h...
WorldMirror 2.0 lets you construct 3D from image/video assets
HY-World 2.0 and WorldMirror 2.0 WorldMirror 2.0 is 1/2 of the recently released HY-World2.0 world generation model framework, which cons...
Lingbot-Map doesn't just generate 3D, it Streams it directly from a video.
LingBot-Map is one of those extremely cool projects that you gotta try to really get it---It doesn't just generate, but STREAMS 3D from a...