Diffusion for Text: Why Mercury Could Make LLMs 10x Faster

Diffusion for Text: Why Mercury Could Make LLMs 10x Faster

Diffusion models changed how we generate images and video—now they’re coming for text.


In this episode, we sit down with Stefano Ermon, Stanford computer science professor and founder of Inception Labs, to unpack how diffusion works for language, why it can generate in parallel (instead of token-by-token), and what that means for latency, cost, and real-time AI products.


We talk through:

  • The simplest mental model for diffusion: generate a full draft, then refine it by “fixing mistakes”

  • Why today’s autoregressive LLM inference is often memory-bound—and why diffusion can shift it toward a more GPU-friendly compute profile

  • Where Mercury wins today (IDEs, voice/real-time agents, customer support, EdTech—anywhere humans can’t wait)

  • What changes (and what doesn’t) for long context and architecture choices

  • The real-world way to evaluate models in production: offline evals + the gold-standard A/B test

Stefano also shares what’s next on Mercury’s roadmap—especially around stronger planning and reasoning for agentic use cases.


Try Mercury + learn more: inceptionlabs.ai


For more practical, grounded conversations on AI systems that actually work, subscribe to The Neuron newsletter at https://theneuron.ai.

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(115)

AI Can Now Write DNA. What Does This Promise, And What Could Go Wrong? With Eric Nguyen, co-founder and CEO of Radical Numerics

AI Can Now Write DNA. What Does This Promise, And What Could Go Wrong? With Eric Nguyen, co-founder and CEO of Radical Numerics

AI can now read DNA like language, generate working biological sequences, and help scientists identify disease-related mutations. Radical Numerics CEO Eric Nguyen explains how genome language models w...

12 Aug 51min

BONUS: How to Use AI Agents for Total Beginners: A Crash Course w/ Agent Builder James McAulay

BONUS: How to Use AI Agents for Total Beginners: A Crash Course w/ Agent Builder James McAulay

This is the #1 request we get every week: how to actually use agents to save time in your business.We’re bringing in James McAulay, founder of The Agent Accelerator, for a practical crash course on wh...

7 Aug 1h 59min

AI Agents Are About to Move Off the Cloud. Here’s What Changes

AI Agents Are About to Move Off the Cloud. Here’s What Changes

What happens when the most capable AI lives in the cloud, but the work you want it to do is too private, expensive, or repetitive to send there every time?Dr. Olena Zhu, Head of AI Solutions & Ecosyst...

5 Aug 54min

BONUS: How to Publish Your AI-Built App to the App Store

BONUS: How to Publish Your AI-Built App to the App Store

Update! The definitive follow-along walkthrough guide is ready! Check it out here: https://theneuron.ai/explainer-articles/how-to-publish-an-ai-built-iphone-app-to-the-app-store/Building an app with A...

31 Jul 1h 8min

AI Can Build It, But Can AI Ship It?

AI Can Build It, But Can AI Ship It?

AI can now help almost anyone turn an idea into a working prototype. But what happens when that vibe-coded experiment needs to become secure, reliable software that real customers can trust—and pay fo...

29 Jul 23min

BONUS: We Built 15+ Apps With AI. Here’s What Actually Worked

BONUS: We Built 15+ Apps With AI. Here’s What Actually Worked

Join Grant Harvey and Corey Noles live on Thursday, July 23, for a show-and-tell featuring the apps, games, extensions, and delightfully specific tools they’ve built with AI, plus tips to help you bui...

24 Jul 2h 20min

Will AI Robots Perform Surgery Better Than Humans?

Will AI Robots Perform Surgery Better Than Humans?

Autonomous surgery may sound futuristic, but automation already plays an important role in procedures requiring extraordinary precision. Mathias Unberath of Inner Logic joins Grant Harvey to explain h...

22 Jul 1h 10min

BONUS: AI Agents Are Here. Now What?

BONUS: AI Agents Are Here. Now What?

Neuron Live is back and we are going deep on Microsoft’s big move in AI agents. 🤖This week, we’re joined by Bryan Goode, Corporate Vice President of Business Applications Marketing at Microsoft, to u...

17 Jul 2h 58min

Populært innen Teknologi

lydartikler-fra-aftenposten
tomprat-med-gunnar-tjomlid
teknisk-sett
shifter
elektropodden
rss-ai-forklart
rss-alt-som-gar-pa-strom
rss-digitaliseringspadden
energi-og-klima
hans-petter-og-co
rss-bak-skyen
fornybaren
smart-forklart
rss-grenser-for-ki
sosialt-sett-om-teknologi-kommunikasjon-og-livet-i-mellom
rss-teknologioptimistene-energibransjens-it-podcast
kortslutning
teknologi-og-mennesker
rss-snakk-om-sikkerhet
rss-ki-til-kaffen