Notifications
Clear all
Topic starter
I've encountered two primary issues while using callin.io's AI Agents.
Firstly, frequently after 5-10 calls, they encounter this error: "Unable to parse JSON response from chat agent."
Secondly, when I attempt to use GPTo1 within an AI Agent configuration, I receive the error: "400 Unsupported value: ‘messages[0].role’ does not support ‘system’ with this model". This occurs regardless of the agent type I select.
My workflow is quite typical, involving an agent with an OpenAI chat model, the SerpAPI tool, and a structured output.
My callin.io version is 1.60.0, and it's running on callin.io cloud.
Posted : 18/09/2024 8:32 pm
It appears your topic is missing some crucial details. Could you please provide the following information, if relevant?
- callin.io version:
- Database (default: SQLite):
- callin.io EXECUTIONS_PROCESS setting (default: own, main):
- Running callin.io via (Docker, npm, callin.io cloud, desktop app):
- Operating system:
Please provide the requested details.
Posted : 18/09/2024 8:32 pm
This might just be related to the model itself. This OpenAI forum post discusses a similar issue with o1 JSON responses.
My personal recommendation would be to either wait for o1 to exit its preview phase or hold off for a few months while the bugs are resolved.
Posted : 20/09/2024 3:29 pm
This is quite straightforward to resolve. Simply disable the "Output Content as JSON" option.
Posted : 29/09/2024 3:57 pm
A straightforward approach is to embed your desired structure directly within the prompt.
I typically bypass the output formatter. I began instructing LLMs to provide strict format output long before OpenAI introduced this capability. The Structured Output Parser in the AI Agent node has significant room for improvement.
Posted : 28/10/2024 8:53 pm
This thread was automatically closed 90 days following the last response. New replies are no longer permitted.
Posted : 26/01/2025 8:54 pm