If Claygent is acting up, it's usually due to processing delays or accuracy issues. Try making your prompts more specific, breaking large requests into smaller batches, and checking your usage limits.
Recent updates to Claude Sonnet 3.7 have made token usage more conservative but more stable. You can also try switching to GPT-5 in the model dropdown - it's showing 8-17% better research accuracy with fewer tokens. [1]
What specific issues are you seeing with Claygent?
If you still need help with the issues you're experiencing with Claygent, I'm here to assist. Would you like to provide more details about what you're trying to resolve or understand?
