Guide
Changelog
Structural adjustments to the user manual and changes to API documentation
Changelog
This page records structural adjustments to the user manual documentation and changes to the API documentation, so developers who have already integrated can cross-check.
The documentation describes only currently valid behavior; if it is inconsistent with your historical implementation, defer to this manual and the live results of GET /v1/models.
This Update (Documentation Standardization)
Navigation Structure
The user manual is now organized by usage path, in six sections:
| Section | Contents |
|---|---|
| Quick Start | Quick start, platform feature guide, model marketplace and billing, common tool integrations |
| Platform Overview | Positioning, features, and FAQ of the four platforms |
| API Access | Authentication conventions, API overview, text, image, video, audio, error codes |
| Asset Library | Asset library guide, asset library API, portrait workflow, reference video |
| Web Creation | Creation workspace and Agent, video generation and retry, FAQ |
| Help & Reference | FAQ summary, billing and statements, rate limiting and quota, changelog, glossary |
Top tabs and the sidebar are updated in sync. All existing URLs remain unchanged; links already shared with customers continue to work.
Added Content
- Added the "Help & Reference" section: FAQ summary, billing and statements, rate limiting and quota, changelog, and glossary;
- Added a separate documentation page for the text-to-speech API;
- Completed feature lists, typical workflows, and FAQ for the four platform pages;
- Added node types, common operations, keyboard shortcuts, and a complete creation workflow example to the Canvas page;
- Added audio API and platform overview entries to the documentation index.
Fixes
- Fixed broken links on the Canvas page that pointed to unpublished pages;
- Fixed the video API link on the main documentation entry to point directly to the task creation page;
- Fixed missing audio API entries in the documentation index to keep it consistent with the API overview.
Concept Clarifications
The following behaviors are explicitly documented in the manual; pay particular attention when integrating:
| Topic | Description |
|---|---|
| Settlement on completion | Async tasks settle on completion; submission only accepts the task and does not pre-charge |
| Queries are not billed | Querying task status incurs no charge; only creating a task is billed |
| Duplicate requests are not merged | Submitting the same parameters repeatedly creates new tasks and bills each one; the platform does not merge them automatically |
| Assets only need to be ingested once | Asset IDs stay consistent across models; switching models only requires changing the model field |
| Portrait assets cannot be downgraded | When switching to a model that does not support portrait assets, they are not automatically changed to ordinary reference images |
| Request snapshot permissions | The request content and request JSON are visible only to administrators; this field is not returned when ordinary users query |
How to Confirm Current Behavior
The documentation describes designed behavior; the actual visible scope of models must be determined by live results:
curl "$LOCOPAL_BASE_URL/v1/models" \
-H "Authorization: Bearer $LOCOPAL_API_KEY"The model catalog changes with account authorization, access address, and platform availability. We recommend re-querying when the service starts, when a user switches models, or when you receive "model unavailable"; do not hard-code model names.