Is MiniMax AI Down?

If Prismix shows MiniMax AI as "Operational" but you're having issues:

• Hailuo video generation queue over 30 minutes — during peak usage hours, Hailuo AI's video generation queue can extend beyond 30 minutes. This is a capacity issue, not a service outage. Check the Hailuo queue status on hailuoai.com and plan video generation tasks for off-peak hours (typically late nights UTC) if turnaround time is critical for your workflow.
• abab6.5 API endpoint returns 404 (migrated to MiniMax-Text endpoint) — MiniMax has consolidated older model endpoints under the MiniMax-Text API path. The legacy abab6.5-chat and abab6.5s-chat model slugs now return 404 on the old endpoint. Update your API calls to use MiniMax-Text-01 or the latest MiniMax-Text model slug via the current API base URL.
• Voice synthesis producing robotic output (speaker_id not set correctly) — MiniMax's text-to-speech API defaults to a generic voice when speaker_id is not specified or is set to an invalid value. The result can sound flat or robotic. Specify a valid speaker_id from the available voice list in the MiniMax documentation, or use voice cloning with a reference audio sample for a more natural output.
• API key from minimax.io not working with hailuoai.com (separate products) — MiniMax (minimax.io) and Hailuo AI (hailuoai.com) are separate consumer and developer products from the same company but use independent account and authentication systems. An API key issued by the MiniMax developer platform cannot authenticate requests to the Hailuo AI web interface, and vice versa. Create accounts separately on each platform as needed.
• Video output resolution lower than requested (plan limitation) — MiniMax video generation enforces resolution caps by subscription tier. Free and starter plan users may be limited to 720p output even when requesting 1080p. Verify your current plan's resolution cap on the MiniMax pricing page and upgrade if higher-resolution video output is required for your use case.
• API rate limit on free tier (2 concurrent requests) — the MiniMax free tier permits a maximum of 2 concurrent API requests. Sending more than 2 simultaneous calls returns a 429 rate limit error. Implement a request queue in your application that limits in-flight requests to 2 at a time, and add retry logic with exponential backoff for 429 responses.