My agent responds to questions slowly on LiveKit c...
# ask-ai
c
My agent responds to questions slowly on LiveKit cloud. When I run it locally, it’s really fast. Is it because I have the Build plan?
t
Thanks for reaching out, Filip. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
Thank you for reaching out! I understand that the slower response time you're noticing on LiveKit Cloud with the Build plan can be a bit frustrating. This usually happens because of agent cold starts—on the Build plan, agents are shut down after each session, so when a new session starts, it can take around 10–20 seconds to get them up and running again. This is why you don’t see this delay when running locally, where the agent stays active all the time. If you’re looking for a smoother, quicker experience, upgrading to a paid plan like Ship or Scale could really help. These plans keep agents warm and ready to go, so you won’t have to wait for those cold starts. If having low-latency responses is important for your work, I’d definitely recommend considering an upgrade. Please feel free to ask if you’d like more details or help with this! Sources: Docs::Home | Quotas and limits | LiveKit Docs | LiveKit Pricing | Builds and Dockerfiles | LiveKit Docs | Builds and Dockerfiles | LiveKit Docs
c
The agent is already up and running. When I’m talking to the agent, it is slower to respond than on my local machine when I run “uv run agent.py console”
Is my phone provider, Telnyx, adding latency?
t
I'd love to better understand how you're using Telnyx in your LiveKit setup. Are you using it for SIP trunking, phone number provisioning, or maybe something else? Knowing this will really help us figure out if it might be affecting the latency you're experiencing. Thanks for sharing!
c
SIP trunking
t
Thank you for reaching out! While there isn't specific information confirming whether Telnyx SIP trunking adds latency to LiveKit Cloud agents, I want to share that LiveKit is thoughtfully designed to keep latency as low as possible using its global edge network for voice and video transmission. When using secure trunking protocols like TLS and SRTP with Telnyx, there might be some impact on performance, but the details aren't clearly outlined. If you're noticing latency and wondering whether Telnyx might be a factor, a good next step would be to run some performance tests or keep an eye on latency metrics during your calls. I'd be more than happy to guide you through how to measure and optimize latency in your LiveKit setup—just let me know! Sources: Community & Support | livekit/livekit_composite | DeepWiki | Getting Help | livekit/livekit_composite | DeepWiki | Create and configure Telnyx SIP trunk | LiveKit Docs | Integrating with Telephony | LiveKit Docs | Agents Overview | LiveKit Docs