Skip to content

Fix ERNIE-Image-Turbo diffusion model link pointing to base model#962

Closed
lin-bot23 wants to merge 9 commits intoComfy-Org:mainfrom
lin-bot23:fix-ernie-turbo-links
Closed

Fix ERNIE-Image-Turbo diffusion model link pointing to base model#962
lin-bot23 wants to merge 9 commits intoComfy-Org:mainfrom
lin-bot23:fix-ernie-turbo-links

Conversation

@lin-bot23
Copy link
Copy Markdown
Contributor

Fixes #961

Changes

In tutorials/image/ernie-image/ernie-image.mdx, the ERNIE-Image-Turbo section had two issues:

  1. The first model download Card under "ERNIE-Image-Turbo model downloads" pointed to ernie-image.safetensors (base model) instead of ernie-image-turbo.safetensors
  2. The Turbo section's "Model storage location" code block also listed ernie-image.safetensors instead of ernie-image-turbo.safetensors

Both have been updated. The base ERNIE-Image section was not modified.

linmoumou added 9 commits April 27, 2026 00:19
… Magnific, PixVerse, Quiver, Veo, Topaz, HitPaw, WaveSpeed

Each provider gets a dedicated mdx page with overview, key capabilities,
and available workflows (Cloud Run + Download JSON cards).
Follows the existing Seedance 2.0 page format.
Split each single provider .mdx into overview + per-model subpages:
- ElevenLabs: overview + 6 model pages (multilingual-v2, scribe-v2, sts, sfx, dialogue, isolation)
- Grok: overview + grok-imagine + grok-video
- Vidu: overview + q1/q2/q3
- MiniMax: overview + minimax-01 + minimax-s2v-01
- Magnific: overview + upscale + creative-tools
- PixVerse: overview + video
- Quiver: overview + svg
- Veo: overview + veo-2-0 + veo-3-1
- Topaz: overview + image + video-enhance
- HitPaw: overview + image-enhance + video-enhance
- WaveSpeed: overview + upscale + flashvsr

New structure: overview.mdx (platform intro + model index) + per-model .mdx (capabilities + workflows)
…, resolution support, style specialties

- Remove 'beta release' phrasing, reflect Pro/Standard/Beta coexistence
- Add ~4s/image speed benchmark, 1K/2K resolution info, image edit input limits
- Add video generation time estimates (~60s extend, ~100s 720p reference)
- Add style specialties: anime, cyberpunk, motion graphics, world rebuilding
…character coordination, dynamic rendering improvements

- Expand camera control details (push/pull/orbit/follow/close-up)
- Add expression & micro-movement, dynamic rendering capabilities
- Add multi-character coordination and scene composition
- Note Turbo/Pro/Standard model variants
- Remove 'Four generation modes' → 'Six generation modes'
…n, duration/amplitude details

Borrowed context from wavespeed.ai model page for parameter details.
…-turbo)

The ERNIE-Image-Turbo section's first model download Card pointed to
ernie-image.safetensors (the base model) instead of
ernie-image-turbo.safetensors. Updated the Card href, title, and the
corresponding path in the Turbo section's model storage code block.

Fixes Comfy-Org#961
@lin-bot23 lin-bot23 closed this May 2, 2026
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.

ernie image turbo link leads to basic model file

1 participant