Mark Daoust
Mark Daoust
Fixed: https://github.com/google-gemini/generative-ai-python/pull/189
This is caused because the code generates the bytes to send tries to create a JPEG file, but the image is RGBA. Adding a `.convert('RGB')` before saving it fixes this....
The internal Eng team has made some significant improvements since this was reported. `block_reason.OTHER` is still a problem but has since been improved. It no longer blocks "who are you"....
It might be that streaming with `transport="rest"` is broken. Edit: No. Rest async streaming is broken.
https://github.com/search?q=repo%3Agoogle-gemini%2Fgenerative-ai-python+%22Unexpected+type+of+call%22&type=issues
> his error occurs when the chat history becomes too long or large. Probably a token issue. Oh, yes. I've seen that error too, I've raised this internally. You can...
500 can be a lot of things, overloaded servers, uncaught exceptions, in this case it sounds like @pkenjora pinpointed the actual problem, it was throwing a 500 instead of "invalid...
`count_message_tokens` was a palm based interface. We've stopped development on those (use `GenerativeModel.count_tokens`). Closing.
Hey everybody, I think @jpdaigle's original issue is resolved. Aside from `429 Quota exceeded` I haven't gotten any errors. I added `request_options=dict(retry=retry.Retry(timeout=600))` to allow retries with along timeout. The retries...
These intermittent ones are often because the model is overloaded. We're going to start 503 for those, which should trigger automatic backoff and retry by the SDK. Also, you can...