app.nzapp
AppsProjectsReposPullsChatIntegrationsGatewayModelsEvalsToolsDatasetsMCPDeploysPricingBlogDocsAssistantsCharactersArtMusic
Sign inStart building
Agent stack
Cloud coding agentAgents SDKIntegrationsBrowser agentMonitors & auto-agentsSchedulersAgent skillsMCP serversDeep research
Models & API
AI GatewayModel catalogModel evalsModel spacesPlaygroundText to imageImage to 3DText to 3DMusic & SFXAudio editorMedia optimizerAI art & libraryChatAPI referenceSchemaBecome a provider
Compute & hosting
DeploysAddonsPostgres hostinggobed vector searchSite hostingAnalyticsCog GPU hostingRL trainingBuilds & CIWorkersTask queuesDomainsGit hosting
Tools
AI toolsDrawDiffusion canvasLive DrawWriteSheetsArtifactsVideo studioNotebooksDatasets
Learn
DocsBlogEval guidesPrompt libraryCLIAlternativesPapersAI charactersArt gallerySecurityConsulting
Company
PricingEnterpriseSettingsBillingStatusInvestorsCreate accountTerms of ServicePrivacy Policy
app.nzapp.nz

AI agent cloud for coding, deploys, model routing, and research. Built for teams shipping software.

Built in New Zealand by App AI NZ.

Social
X / TwitterGitHubYouTube
The app.nz network
GpuBrainPapersReading TimemojojojoNetwrckText-Generator.ioCodex InfinityOpenPathsCuteDSLAI Art GeneratorAIArt-Generator.artSiteSimSimplexGenDictatorFlowWebFiddleRing.nzChatGibidyBitBankExperimentFlowEvangelerHires.nzHow.nzV5 GamesAddicting Word GamesBig Multiplayer ChessWord SmashingreWord GameMultiplication Master
© 2026 App AI NZ Ltd. All rights reserved.All systems normalTermsPrivacy
Blog
July 22, 2026·8 min read·app.nz

ComfyUI lab 12: Turn a concept image into a 3D asset

Prepare an object reference that communicates shape cleanly, then validate the generated asset outside the preview renderer.

Listen to this article

On-device voice

Uses the voice built into your browser; no article text leaves this page.

Audio narration is not supported by this browser.

Image-to-3D is an inverse problem: one view does not reveal the hidden side. The reconstruction system must infer thickness, back geometry, occluded parts, scale, and material. Your reference should reduce that ambiguity.

Build a reconstruction-friendly image

Use one object, centered, fully visible, with a three-quarter view and a quiet contrasting background. Prefer broad soft lighting that reveals form without crushing shadows. Avoid motion blur, transparent materials, mirrors, extreme perspective, and cropped extremities. For production, multiple calibrated views or a dedicated capture pipeline are better than one beautiful concept.

The graph loads an image, passes the IMAGE value to the MIT app.nz bridge, and saves the returned GLB. Compare its result with open references such as Stable Fast 3D, but compare licenses, hardware, input assumptions, and topology—not just one rotating preview.

Run an asset acceptance test

Measure silhouette overlap against the reference, inspect the unseen side, verify the base sits on a plane, check scale and pivot, and load the GLB in the actual target engine. Decide early whether the asset needs animation-ready topology, collision, LODs, UV editing, or only a background prop.

Reconstruction gives you a candidate mesh. Retopology, material cleanup, and art direction remain normal 3D work.

Run the lab

  • Open this graph in Comfy Studio
  • Download the exact API-format workflow
  • CLI: app comfy download appnz-image-to-3d

Record the graph JSON, input assets, seed where relevant, model or service version, custom-node commit, and final artifact together. That bundle—not a screenshot—is the reproducible creative unit.

Open-source map

  • Open app.nz 3D node — MIT image-to-3D bridge
  • Stable Fast 3D — Open single-image mesh reconstruction project

Creative Systems Labs: 12 of 12. Previous: Design a text-to-3D handoff You completed the lab collection. Combine the audio, character, reactive, style, and 3D graphs into your own production pipeline.

Build what you just read

Ship agents, models, and apps on one cloud.

Start with free credits, then use the same platform from the web app, CLI, desktop app, or MCP.

Start building freeRead the docs

Keep reading

ComfyUI lab 11: Design a text-to-3D handoff

Prompt for geometry rather than beauty, generate a GLB, and inspect silhouette, topology, scale, and material assumptions.

ComfyUI lab 8: Prototype frame transforms safely

Understand image tensor shape, range, batch semantics, and the boundary between a prototype code block and a maintained node.

ComfyUI lab 5: Turn sound into reactive motion

Window a waveform, measure local energy or spectrum, map it to frames, and preserve synchronization at a declared FPS.