Back to all discussions

  •   about 2 months ago

Intermittent timeouts / high latency when querying patient data via agents

I’ve been working on building an agent using uploaded patient documents (PDF reports) and MCP integration. The setup was working correctly earlier, but I’m currently experiencing consistent issues when querying the agent.

Issues observed:

Frequent errors like:
“The LLM took too long to respond and the operation was cancelled”
“This model is currently experiencing high demand”
Requests either fail or take too long, even for simple prompts like:
“Give me patient details”
“Analyze thyroid reports and identify trend”

Context:
Hi everyone,

Scope: Workspace
Agent: Custom patient summary agent (also tested with General Chat Agent)
Data: 4 uploaded PDF reports in a collection
This was working previously but has become unreliable

What I’ve tried:

Retrying multiple times
Using shorter prompts
Switching between agents
Refreshing session

Still seeing the same issue consistently.

Question:
Is this due to current system load or rate limits?
Are there any recommended workarounds or best practices to improve reliability?

Thanks in advance — really enjoying the platform otherwise and would love to continue building on it.

  • 1 comment

  • Manager   •   about 2 months ago

    Hi Pallavi - unfortunately, there is not much we can do here. This is an error from the model provider. We have a default of 60 seconds for the timeout. We are reviewing a feature to allow users to configure the timeout. You could try creating a new account on Gemini. We are also having some discussions with other and might have some other solutions based on the feedback from one of the major model providers. I hope that was helpful. Please let us know if you have any other questions. You can also join the discord channel and ask them there

Log in or sign up for Devpost to join the conversation.