Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.speckle.systems/llms.txt

Use this file to discover all available pages before exploring further.

What are models?

Models are containers for design data inside a project. Use separate models to keep ownership, review cycles, and version history clear. This page is for project owners and discipline contributors deciding how to split project content for day-to-day publishing.

Why model boundaries matter

Model boundaries affect review clarity and troubleshooting speed. When one model mixes unrelated scopes, teams struggle to identify who owns changes and which versions should be promoted. Use separate models when teams differ by discipline, area, package, or release schedule. Keep related content together when people review and publish it as one unit.

Split into separate models

Use when ownership, discipline, package, or release schedule differs.

Keep in one model

Use when same team reviews and publishes changes as one unit.

Create a model

A new project can start with no models. Create a model first, then publish versions into it. Use a model as a stable container. Versions carry the evolving design snapshots over time.

Organize model names

Use model names that match your team structure. You can use / in names to group related models, for example Building A/Floor 01. Prefer naming conventions that help new team members predict where data should live. If your team cannot agree where to publish, your naming structure is too ambiguous.

Work with versions

Clicking a model opens that model’s home page. Use this page to review model-level activity and publish new versions before opening specific version details.

Model home page

The model home page surfaces model-level signals you need to coordinate work:
  • A current snapshot preview of the latest version.
  • A drag-and-drop upload area to publish a new version directly.
  • Version history for that model.
  • Upload history for recent imports and publish events.
  • Issues linked to that model.
  • Data Validation KPIs for that model.
Some sections are dynamic and appear only when that content exists (for example, Issues or Data Validation results). For current validation guidance, see Validate your data in Speckle and Viewing Results.

Next step

Read Versions for history and version workflows after first publish.

FAQ

No. Project labels are a workspace-wide tag set for projects (see Project labels). You cannot attach that same label system to models inside a project today. Use model names and path-style groupings with / (see Organize model names) so models stay easy to scan in the list.
Not always. Split by discipline when ownership and release timing differ. Keep one model when the same team reviews and publishes together.
Yes, but renaming late can break team habits and external references. Align naming early to reduce migration work.
No. Access is controlled at project and collaborator level. Models help structure content inside those access boundaries.
Create a new version for change over time in same scope. Create a new model when ownership or scope boundary changes.
Last modified on May 4, 2026