Pinhole: a one-button fully local image generator on stable-diffusion.cpp, 99% written by AI agents
Pamidoraa · reddit · 2026-10-01
A developer released Pinhole, a fully local image generator with no node graphs, no Python, and no slider overload: pick a model, type a prompt, hit Generate. It detects your GPU, suggests fitting models, and auto-configures text encoders, VAE, steps, and samplers; supports describe-to-edit and image-to-prompt, with everything visible in a Fine-tune drawer.
- Model support: Qwen-Image 2.1 default on 12GB+ (also for editing), Z-Image Turbo on 6-10GB, SDXL finetunes on smaller cards, SD 1.5 on CPU; also SDXL/Pony/Illustrious, FLUX.1, FLUX.2 klein/dev, Chroma, SD3/3.5, HiDream, LoRAs.
- Built-in CivitAI browser (safetensors/GGUF only) and can reuse existing ComfyUI/Forge model folders.
- Everything stays on-device, no account or subscription; Windows/Linux, NVIDIA/AMD/Intel; local safety safeguards built in.
- Engine is a fork of stable-diffusion.cpp; 99% of the code was written by swarms of AI coding agents, with the author steering.
More from Multimodal
- AutoTrust AI debuts JEV-27B-VL, an open-weight multimodal decision model that acts on what it sees — alexcovo_eth · 2026-10-01
- Mixie3D launches: AI agents model, texture and light 3D scenes in parallel — gowthami_s · 2026-10-01
- A Qwen outpaint LoRA beats native outpainting on quality and consistency — linoy_tsaban · 2026-10-01
- Midjourney Artist Opens Her Entire Archive and Site mjpro.ai for Free — ciguleva · 2026-10-01
- Seeddance 2.5 used to create a 3D animated Louvre heist scene, prompt shared — azed_ai · 2026-10-01
- Orbit LoRA plus first/last frames freezes time in MiniMax video with 360 camera moves — AndrewJumpen · 2026-10-01