Incredibly Slow Agent Turns
Hey team! Just wondering if anyone has tips for finding what is slowing down my agent?
Using structured working memory with gpt 5 mini. Working on finding the right reasoning effort but wondering if the slowdown could be coming from other parts of the agent loop. Has anyone worked on speeding up their mastra agents?
I have agent turns running upwards of 20 minutes according to the ai_spans started and ended diff
4 Replies
when i started with Mastra it happen to me as well that I had runs for ~15-20 min long.
I didnt figure it out where it did coming from, but I was suspecting it happend when I added memory, semantic search and embeddings.
📝 Created GitHub issue: https://github.com/mastra-ai/mastra/issues/8959
GitHub
[DISCORD:1428523877528506428] Incredibly Slow Agent Turns · Issue ...
This issue was created from Discord post: https://discord.com/channels/1309558646228779139/1428523877528506428 Hey team! Just wondering if anyone has tips for finding what is slowing down my agent?...
Hey @Chad Holmes ! It's hard to tell what is making it slow. How big is your context? Do you have a way for us to reproduce this issue?
I figured it would be a bit hard to test. Context size is ~100k input tokens. I noticed that the agent was looking at past messages and reduced that and it sped up quite a bit. But just hard to trace which steps are slow. Is there somewhere in the trace I can see the actual network latency?