[lin-bot23]

@lin-bot23

Adds the missing docs for the templates_mjm_image_to_3d workflow, which is live in Comfy Cloud and in the template library but had no mention anywhere in the docs.

What is documented

A new section, Single-Image to Model with Generated Views, on tutorials/partner-nodes/tripo/model-generation.mdx (en, zh, ja, ko, es):

• One photo in, 3D model out. Nano Banana 2 removes the background and renders a side view and a rear view, all three images are saved, and the Tripo image-to-model node builds the model from the background-removed photo.
• Cloud card and workflow download card, plus the sample input image.
• What one run costs in API calls (three images and one model) and where the GLB lands (ComfyUI/output/3d/).
• The part users cannot guess from the preview: the mesh comes from the background-removed photo alone, because the node accepts one image and takes the first frame of the batch. The generated views are PNG outputs, and the section points to the multi-view workflow for models built from several views.

Korean page restoration

The Korean page was missing the Cloud card, download card and input card for the Image-to-Model workflow (it still used the old raw link button). Restored to match the English structure, with the same hrefs.

Fact sources

• workflow_templates/templates/templates_mjm_image_to_3d.json: node list and wiring. BatchImagesNode.image0 is the background-removed original (link 28) and feeds TripoImageToModelNode.image (link 5); the side and rear frames are slots 1 and 2 and are only wired to SaveImage. Three GeminiNanoBanana2 nodes, Preview3D, and SaveGLB with prefix 3d/tripo_image2model.
• comfy_api_nodes/nodes_tripo.py: TripoImageToModelNode takes a single image input and calls upload_images_to_comfyapi(cls, image, max_images=1), which is why the generated views do not reach the API. TripoMultiviewToModelNode is the node that accepts separate front, left, back and right images.
• comfy_api_nodes/nodes_gemini.py: the Nano Banana 2 display name.

Verification

• All URLs in the new sections return 200 (preview webp, input image, workflow JSON, Cloud link).
• check-structure-parity.ts --base=upstream/main: OK. The remaining Korean h3 difference is pre-existing drift and is reported as such.
• check-anchors.py on the five page files: OK.
• 6 ## headings and 3 ### headings in every locale, identical CardGroup/Card/img counts, no em dashes, no leading pipes.

Note for reviewers: templates/index.json lists Meshy under models for this template and its description says the views are used with Meshy, while the template JSON uses the Tripo node. The page follows the actual workflow. The template also ships the legacy Tripo image-to-model node (is_deprecated=True) and batches three images into a single-image input, so the side and rear views are generated but unused. Worth fixing in the template rather than in the page.

[mintlify]

Preview approval required

This pull request comes from a fork, so Mintlify won't build a preview until someone with write access approves commit cb39eea.

Approve preview

[coderabbitai]

No actionable comments were generated in the recent review.

Walkthrough

The tutorial adds a single-image workflow that generates side and rear views, then creates a 3D model from the background-removed photo. Localized pages document the workflow, its links, API calls, and output files.

Changes: Tripo workflow documentation

Layer / File(s)

Summary

Document the generated-views workflow tutorials/partner-nodes/tripo/model-generation.mdx, es/tutorials/partner-nodes/tripo/model-generation.mdx, ja/tutorials/partner-nodes/tripo/model-generation.mdx, ko/tutorials/partner-nodes/tripo/model-generation.mdx, zh/tutorials/partner-nodes/tripo/model-generation.mdx

The pages describe background removal, generated side and rear views, template and sample-image links, four API calls per run, and GLB and PNG outputs. They state that the mesh uses the background-removed photo and direct readers to the multi-view workflow for models using several views. The Korean page also changes the existing image-to-model section presentation. Translation metadata tracks the new section.

Possibly related PRs

• Comfy-Org/docs#2101: Removes Spanish locale navigation and translation configuration, which relates to the Spanish tutorial’s availability.

Suggested reviewers: robinjhuang

Merge Risk: Minimal · up to cb39e

The documentation explains that only the background-removed photo builds the mesh and that generated views are saved separately. No concrete merge-blocking issue is established; merge after normal documentation checks.

Pre-merge checks: 2

Check name	Status	Explanation
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Finishing Touches

Generate unit tests (beta)

• Create a new PR

Simplify code

• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Recent review info

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: ASSERTIVE
• Plan: Advanced
• Run ID: 5aa5bc9b-592f-4b73-9049-7b6a663fe03b

Commits

Reviewing files that changed from the base of the PR and between 4a9f856 and cb39eea.

Files selected for processing: 5

• es/tutorials/partner-nodes/tripo/model-generation.mdx
• ja/tutorials/partner-nodes/tripo/model-generation.mdx
• ko/tutorials/partner-nodes/tripo/model-generation.mdx
• tutorials/partner-nodes/tripo/model-generation.mdx
• zh/tutorials/partner-nodes/tripo/model-generation.mdx

Included review availability: This review used your included allowance. Your plan provides up to 2 included reviews per hour; 1 remain after this review.