Daily.co

Daily.co

Real-time voice, video, and AI for developers

4.9
27 reviews

639 followers

Build live video and audio applications using modern APIs and end-to-end tooling.

This is the 10th launch from Daily.co. View more
Daily Bots

Daily Bots

Build voice-to-voice AI agents with any LLM with Open Source
Build hosted AI agents on Daily Bots, the Open Source cloud for ultra low latency voice and video. Use commercial and open models. Incorporate tool use, telephony, vision, etc. If your needs evolve beyond Daily Bots, run your code anywhere that supports RTVI.
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Daily Bots gallery image
Payment Required
Launch Team / Built With

What do you think? …

Kwindla Kramer
Hi Product Hunt! Today we’re sharing Daily Bots, a hosted platform for real-time voice and video AI bots. We built Daily Bots to help developers ship and scale voice and video agents, using Open Source SDKs and deployed on Daily’s global infrastructure. For the past 18 months, we've been experimenting with and building tools for voice-to-voice and real-time video AI. We've helped our customers ship games, meeting co-pilots, medical documentation systems, customer support agents, virtual interviewers, teaching assistants, low-code voice AI platforms, virtual avatar infrastructure, and much more. We love this new world of generative AI and we want to make real-time AI accessible to every hacker, maker, and product team. We've helped create two big Open Source toolkits for real-time AI. Pipecat enables server-side "orchestration" of multi-model, multi-modal AI workflows. And RTVI is an open standard for connecting clients to real-time voice and video inference backends. Daily Bots is built on Pipecat and RTVI. You can build a Daily Bots application using the Open Source RTVI SDKs for Web, iOS, and Android; ship fast and scale quickly on Daily's global WebRTC infrastructure; and retain the flexibility to move to a different platform or run your own infra if you want to in the future. With Daily Bots, you can create AI agents that talk naturally, with fast voice-to-voice response times, interruption support, and multi-turn context management. You can use both commercial and open models. We’ve partnered with Anthropic, Cartesia, Deepgram, and Together AI. (And you can use any other LLM that supports OpenAI-compatible APIs.) Finally, you can buy phone numbers from us and configure your bot to accept dial-in telephone calls. We're excited to see what you build. Let us know what you're thinking about making with Daily Bots. And if you just want to experiment with a few LLMs, prompts, and voices, check out this demo: https://demo.dailybots.ai/
Aarav Krishna
I love the flexibility this offers. How does it handle integrating with existing telephony systems?
Chad Bailey
@aarav_krishna We already have an integration in place that makes it pretty straightforward to use Twilio voice. If you're using a different service that speaks websockets, we could probably help you make the right connections in a Pipecat bot that runs on Daily Bots or on your own infra.
varun
@aarav_krishna We offer PSTN phone numbers from which you can dialout or receive dialin. Similarly, we offer SIP interconnect which means that you can connect any third-party like twilio, telnyx. For example if you have phone numbers on third-party, you can forward the twilio call to Daily's SIP Interconnect (sip_uri exposed for your domain) and everything should automatically work. See https://docs.dailybots.ai/api-re... for PSTN dialin, SIP would be similar, just skip step 1.
Kyrylo Silin
Hey, I'm wondering what some of the most interesting or unexpected use cases you've seen so far have been? For developers looking to get started, what would you say is the learning curve like for creating a basic voice bot using your platform? Congrats on the launch!
Kwindla Kramer
@kyrylosilin Daily Bots isn't a low-code tool. We're aiming at developers who want to build on Open Source and understand some of the architectural details. I think Vapi (https://vapi.ai/) has the best combination of powerful capabilities and gentle learning curve of any of the voice AI platforms. Having said that, a Daily Bots "hello world" voice app in JavaScript is less than 20 lines of code. I think the most surprising thing to me about the voice AI space is how many people are building their own private voice agents. It feels like a new frontier for personal project hacking. Build your own virtual assistant/co-pilot/therapist! I also think there's going to be a whole new generation of voice-enabled hardware that works really well and that we all take for granted within a few years.

Do you use Daily.co?