Skip to content

Add Video Enhancement guide for the Topaz suite - #469

Open
tombystrican wants to merge 4 commits into
mainfrom
tomb/topaz-api-docs
Open

Add Video Enhancement guide for the Topaz suite#469
tombystrican wants to merge 4 commits into
mainfrom
tomb/topaz-api-docs

Conversation

@tombystrican

@tombystrican tombystrican commented Aug 22, 2026

Copy link
Copy Markdown
Contributor

Summary

Tracks https://linear.app/venice/issue/VEN-6530/topaz-v2-video-enhancement-suite-on-fal-9-models-full-settings

Documentation for the Topaz video enhancement suite (ships with https://github.com/veniceai/outerface/pull/10024 and https://github.com/veniceai/interface/pull/12949).

New: guides/media/video-enhancement.mdx

Full guide for all nine Topaz task models (Precision, Starlight, Starlight Fast, Astra, Hyperion, Interpolate, Denoise, Deblur, Colorize):

  • Model table with IDs and what each does
  • Complete settings reference: upscale_factor, enhancement_model (full per-model option lists), all sliders with ranges, target_fps, slowdown_factor, output_format, h264_output, and Astra's guidance prompt
  • Per-tier pricing table (≤720p / 1080p / 4K per second) including the per-model rate differences (Gaia 2 / Nyx Fast half price, Aion ~2.7x, Proteus Natural 4K discount) and the fps rules (2x at ≥48fps on upscale endpoints, linear scaling on Interpolate)
  • Constraint callouts: Nyx Fast/HF are denoise-only, Astra prompts cap input at 450 frames, H.264 limited to 4K, Astra snaps its own output resolution
  • Per-task curl examples (restore, denoise, SDR→HDR ProRes, slow motion)
  • Discovery via GET /modelsconstraints.enhancement

Updated

  • guides/media/video-upscaling.mdx — deprecation warning for topaz-video-upscale, pointing to topaz-video-precision as the direct replacement
  • api-reference/endpoint/video/queue.mdx — the upscaling section now covers the enhancement suite and its per-model settings
  • models/video.mdx — model types mention the enhancement suite
  • swagger.yaml — synced from outerface (enhancement params + constraints.enhancement discovery block)
  • docs.json — nav entry after the upscaling guide

Merge order

Merge after the outerface PR is deployed and the models are flipped from staff-only to public (GA) — the synced swagger and the new guide document the public API surface. The models are staff-only until GA.

New guides/media/video-enhancement.mdx covering all nine Topaz task models
(precision, starlight, starlight-fast, astra, hyperion, interpolate,
denoise, deblur, colorize): full settings reference (upscale_factor,
enhancement_model, sliders, target_fps, slowdown_factor, output_format,
h264_output, astra prompt), per-tier pricing table, constraints and
per-model validation notes, and per-task curl examples.

Also:
- video-upscaling.mdx: deprecation warning for legacy topaz-video-upscale,
  pointing at topaz-video-precision as the direct replacement
- queue.mdx: the upscaling section now covers the enhancement suite and
  its per-model settings
- models/video.mdx: model types mention the enhancement suite
- swagger.yaml: synced from outerface (adds the enhancement params +
  constraints.enhancement discovery block)

Depends on the outerface PR that ships the models — merge this after
veniceai/outerface#10024 is live.
@mintlify

mintlify Bot commented Aug 22, 2026

Copy link
Copy Markdown
Contributor

Preview deployment for your docs. Learn more about Mintlify Previews.

Project Status Preview Updated (UTC)
veniceai 🟢 Ready View Preview Aug 22, 2026, 12:33 PM

Drops the deprecation warnings — the legacy model stays live with no
retirement signaling until the replacement decision is made. The
upscaling guide now points to the enhancement suite as a tip.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant