open-higgsfield
What they did
An open-source studio with one prompt bar, per-model settings and a single gallery of finished runs.
Worth knowing
- Price: Open source repo (license unknown)
- Status: sourced. We have not run this ourselves yet (no "Tested" badge).
- Quality bar (computed by code): 4/5 (primary source ✓, reproducible ✗, new ✓, example output ✓, safe ✓)
The best new AI video workflows and prompts, weekly.
From the source
Setup. Clone and self-host it (Next.js 16, React 19, pnpm, deployable on Vercel), or use the hosted studio at openhiggsfield.ai. Either way you add your own platform key (id:secret). The studio itself is free; you pay the model provider.
Inputs. One prompt bar for both image and video across 38 models (8 image, 30 video), including Seedance 2.5/2.0, Kling 3, Wan, LTX, MiniMax, PixVerse, Grok and Flux. Per-model settings (aspect ratio, resolution, duration, audio, batch up to 4), plus start frame, end frame, reference, video and audio inputs by role.
What the source says about results. Every finished run lands in a masonry gallery (Image, Video, Assets, Favorites). 'Reuse' restores the exact model, settings and prompt of an earlier run. The README positions it as an open-source alternative to Higgsfield's studio subscription.
Summarised from github.com/wide-trace/open-higgsfield, read 2026-10-11. Claims are the source's, not our tests.