Added on Nov 19,2026

uRun is an AI inference cloud for voice and video agents that need to react in real time. Its flagship model, Sense, is a perception model that answers questions about live audio, video and screen streams while the conversation is happening.
Sense returns answers 12.5 times a second, always about the current moment. Developers write the questions in plain language, such as whether the user has finished speaking, whether they just interrupted or whether the agent should stop talking, and the agent acts on the structured answers.
Answers come back in three forms: yes or no, a choice from a list, or a score. Audio and video can be handled in one stream, and different sets of questions can give the same agent different turn-taking personalities.
Example uses listed on the site include turn-taking for voice agents, support escalation when caller frustration passes a threshold, live moderation, guided tasks over video, camera security alerts and monitoring of telemetry streams.
Developers build with an SDK for real-time sessions or send single requests to explore the model. uRun also offers custom model serving, where teams run their own weights or pipelines on GPUs such as H100s with help from uRun engineers.
Pricing: usage is billed in credits per stream-minute, with Sense at 80 credits per stream-minute. The Basic plan is 20 US dollars a month for 10,000 credits and the Team plan is 150 US dollars per member a month for 120,000 credits, while Scale is custom. A free tier exists and new accounts receive launch credits.
It is suited to developers and AI teams building voice assistants, video agents, call-centre automation and other real-time multimodal applications.
By the PowerfulAI.tools Editorial Team · Updated 20 November 2026 · Researched with AI assistance and checked against the official uRun website · How we review tools
uRun is freemium: free to start, with paid plans for more usage and features. Check uRun's website for current prices.
Developers building live voice agents, speech translation or real-time video experiences who need low-latency hosted model APIs.
Whisper large-v3 on uRun handles 99 languages with translation into English text, while Kokoro 82M offers streaming text-to-speech in nine languages.
Yes. Model labs can run their own models or pipelines on uRun and work with its engineers on configuration, and distribution to other developers is opt-in.
Popular alternatives to uRun include Pollinations, NemoClaw and Agnes AI. Compare their pricing and what each is best for below.
| Tool | Pricing | Best for |
|---|---|---|
| Pollinations | Freemium | Developers and hobby builders who want a single API for adding AI images, audio and text to their websites, bots and app... |
| NemoClaw | Free | Developers and platform teams who want to run always-on AI agents on NVIDIA hardware under policy-based security control... |
| Agnes AI | Freemium | Developers building AI-powered apps who want access to several models through one API. |
| OrcaRouter | Freemium | Developers and teams who want to balance AI quality and cost across many models. |
| DiagramGPT | Free | Developers and students who need to sketch system designs and workflows quickly by describing them in words. |
Leave a review for the community
No Review Found
Already a member? The same buttons sign you in.
We'll email you a 6-digit code. No password needed. By continuing you agree to our Privacy Policy.
We sent a 6-digit code to . It expires in 10 minutes.
No email? Check your spam folder, or ·