why is figma weave partially offline

A partial outage occurs when core canvas features remain fully functional while cloud-heavy AI microservices experience service degradation. During these events, Figma Weave might allow you to open existing files or view previously generated frames, but sending new prompt requests or syncing design tokens fails intermittently.

Understanding why partial service disruptions happen makes it easier to navigate temporary performance hits without losing work or consuming excess usage quotas.

Quick checks

• Test simple text prompt requests to see if complex multi-frame layouts are the only commands stalling.
• Monitor browser network logs for recurring 502 Bad Gateway or 504 Gateway Timeout responses on API endpoints.
• Confirm whether other team members on different local networks experience similar slowdowns.

1. Identify affected microservices. Generative AI architectures rely on separate pipelines for prompt interpretation, layout rendering, and asset delivery. If rendering stalls while layout wireframes complete, follow /fix/figma-weave-not-generating to re-establish proper generation handling.
2. Check for billing and credit validation delays. High traffic during partial outages can cause background token validation to time out. If you receive sudden payment or entitlement warning banners while the service is lagging, visit /fix/figma-weave-billing-error to diagnose quota synchronization issues.
3. Differentiate between server degradation and local connectivity drops. Intermittent dropouts often signal regional routing disruptions between your ISP and cloud clusters. To evaluate overall system reachability, see /fix/figma-weave-down for detailed network testing protocols.
4. Fall back to manual canvas assembly temporarily. When AI layout generation operates at reduced capacity, avoid submitting rapid, repeated prompts as this increases queue congestion and risks temporary rate limits.

Still stuck?

Figma Weave Support is an independent user resource. If partial service disruption persists across multiple files and accounts, allow backend infrastructure services time to balance traffic loads before resuming large-scale design tasks.