Image-to-video in ComfyUI without a local GPU: Forma custom nodes, pay-as-you-go from $5
no3us · reddit · 2026-10-22
Reddit user no3us shares a subscription-free image-to-video setup for ComfyUI: install the free Forma custom nodes, connect an account, and run cloud inference from local ComfyUI — no local GPU or model downloads required.
Key points:
- 39 image/video models, 8 video workflow types, 2 image workflow types
- Pay as you go with $5 top-ups; quote shown before running
- Payments via cards, Apple Pay, Google Pay, Link, or crypto
Nodes, install instructions, and example workflows are open-sourced on GitHub (vavo/runpod-endpoints-comfy); the demo image shows $0.16 per generation.
More from Infra
- "It's bad out there rn": Cloudflare DE lead urges everyone to put sites behind Cloudflare — threepointone · 2026-10-22
- ATEST: load-aware test-time scaling cuts LLM inference energy up to 48x — Mobina Kashaniyan · 2026-10-22
- Liquid cooling yields ~22% more compute per megawatt, NVIDIA-Vertiv DSX deal shows — BenBajarin · 2026-10-22
- NVIDIA investing billions to help Europe build its own AI factories — karlfreund · 2026-10-22
- When the Entire Internet Went Down, One Homelab Stood Tall — haydendevs · 2026-10-22
- Grace Hopper Celebration canceled over funding; SGLang Summit gives away 20 Expo passes — BanghuaZ · 2026-10-22