uRun
A real-time decision model for voice and video agents — it listens to a live stream and answers the developer's questions up to 12.5 times a second (is the user done speaking, did they interrupt, should the agent stop talking).
Why it matters
Its first model, Sense, is built to walk forward through a live stream and orchestrate turn-taking in real time — the 'when to speak' decision layer that most voice agents currently fake with fixed timeouts.
What you could build with it
A developer building a phone-based support agent could use Sense to detect the exact moment a caller finishes speaking or interrupts, so the agent stops talking mid-sentence naturally instead of waiting out a fixed silence timeout.
Does it hold up?
Too early to judge — the launch post claims 12.5 decisions/second and 1,000 free credits for new accounts, but no independent latency or accuracy benchmarks exist yet.
Built with uRun
- Market News & Intelligencebibamoney.com · Nov 19 news relay noting the uRun Product Hunt launch for real-time decision-making in voice and video agents.
- Product Hunt: today's launchescrosswalk.to · Nov 18 Product Hunt homepage list showing uRun at #39: 'Real-time decisions for voice and video agents'.
- urun-sh/sense-sdkgithub · The maker's own Python/TypeScript SDK for building real-time Sense sessions (named in the launch post).
Learn more
First spotted on article: source.
More AI models
Edge0-145B: open 145B MoE that runs on a single consumer GPUEdge0 open-sourced 145B and 120B latent-MoE models derived from Qwen3.8-Flash-Next, plus an inference framework that…models · JEV 0.74Strata: run a 125B model on a gaming PCFree, MIT-licensed local inference engine that runs Qwen3.8-Flash-Next (125B MoE, about 6B active per token) on…models · JEV 0.73Qwen4-Max: Alibaba's new flagship model goes liveAlibaba's 2.4T-parameter flagship is now generally available via API, with a 1M-token context, native vision, and…models · JEV 0.67smolmaxxing: run 16B LLMs on a 1 GB TV box or a 2013 iPadAn MIT-licensed llama.cpp patch set that runs language models up to a 16B MoE on devices with about 1 GB of usable RAM…models · JEV 0.65Blue TTS v3: a 388MB open text-to-speech model with voice cloningAn open-source 388MB text-to-speech model (MIT) with voice cloning from about 10 seconds of speech, 7 languages, 15…models · JEV 0.65Haru-4BOpen full-duplex spoken-dialogue foundation model that listens and speaks simultaneously (7.5 Hz frames, 133 ms per…models · JEV 0.63
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.