API Platform: Arrow 1.x retires October 16
By Steven Van ·
Requests to arrow-1, arrow-1.1 and other Arrow 1.x IDs return 404 after the cutoff, so API users must switch to arrow-2 and may need new keys.
Four QuiverAI API model IDs stop working at 08:00 UTC (10:00 Berlin) on October 16, 2026, according to the vendor's changelog. The retiring IDs are arrow-1, arrow-preview, arrow-1.1 and arrow-1.1-max. arrow-1.0 is already rejected and stays rejected.
What happens at the cutoff
Until then, Arrow 1.x requests and GET /v1/models behave as they do now. At the cutoff the affected IDs stop serving, disappear from GET /v1/models and return 404 model_not_found. Requests are not silently routed to Arrow 2.
Moving to Arrow 2
QuiverAI says to use arrow-2 for SVG generation, vectorization, editing and animation. Native endpoint paths do not change. Pricing does: Arrow 2 uses measured token pricing instead of the fixed-credit prices used by Arrow 1.x.
API keys
How a key migrates depends on its model authority:
- A key that allows all models only needs the request model changed to arrow-2.
- A key scoped to an exact Arrow 1.x model is immutable. Create a new key for arrow-2, deploy the new secret, then revoke the old key once traffic has moved.
- Rotating a key replaces its secret but keeps its authority, so rotation is not a model migration.
New exact-model keys can no longer select Arrow 1.x. Existing Arrow 1.x-scoped keys keep working until the cutoff.
