Instantiating and/or querying the model results in 404

[samwinslow]

Instantiating and/or querying the model results in 404

[brannondorsey]

Thanks for submitting this report. I can try and help debug here in GH issues. Feel free to reach out to us via the chat bubble at support.runwayml.com too and we can discuss details about your user account that may be contributing to this.

Every time I encounter this error, I find that the hosted model has become Inactive, as shown on the dashboard at https://app.runwayml.com/hosted-models. I can then toggle it to Active, and perhaps make 1 request through my app before the model is once again Inactive.

This is interesting, I haven't seen this before. Are you able to have multiple hosted models "Active" at once? Non-subscribed users can only have one hosted-model active at a time. It's possible there is an issue on our end that is incorrectly marking hosted-models as inactive when it shouldn't.

have been making around 50-60 requests/hour while testing. If rate-limiting is the case here, it would be nice if the model returned 429 Too Many Requests instead of silently inactivating.

As for rate-limiting, we do have a relatively high global rate limit on hosted-models, however that returns a 429 and it sounds like you haven't hit it. Our rate-limiting functionality also wouldn't mark your hosted-model as inactive even if it were triggered.

Like I said, I'm happy to take this up in a more real-time based conversation in our support chat. I can then report our findings back here for other folks to read in the future.

[samwinslow]

Hey Brannon, thanks. I have only one hosted model in my account which is on a paid plan. Will reach out via the live chat.

[samwinslow]

This was resolved via a Support chat and turned out not to be a problem with the Hosted Models library.

[brannondorsey]

For future reference, Hosted Models are automatically deactivated if the user has a $0 credit balance. Topping up the credit balance and re-enabling the Hosted Model should get things working again.

Thanks for raising this issue @samwinslow!