Nvidia says its Groq 3 LPX racks delivered 3,400 tokens per second in an Artificial Analysis benchmark running Gemma 4 31B with a 100,000-token input sequence
THE SO WHAT
3,400 tokens/sec on a 100,000-token Gemma 4 31B prompt is Nvidia signaling that long-context, low-latency inference is moving into production territory. If your workloads are bottlenecked on context length or response time, it’s time to re-run your infra and vendor benchmarks with LPU-class options in the mix.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIOpenAI loses a second sales leader in a week to Salesforce
Two senior go-to-market leaders boomeranging back to Salesforce within a week suggests the frontier-model vendor vs. incumbent-SaaS calculus is shifting for top enterprise sellers. If you’re building on OpenAI or competing with it, assume near-term sales org turbulence and double-check who actually owns your relationship and roadmap commitments.
Applied AIThe biggest question about AI is what humans will do next
The real constraint on AI impact is not model capability but how quickly organizations can rewire roles, incentives, and processes around it. Treat AI adoption less like a tooling upgrade and more like a multi-year recovery plan with setbacks baked in, or you’ll overestimate progress and underinvest in change management.
Applied AINew Apple HomePods with AI are coming. Everything we know.
An AI-first HomePod line — including a screen and tabletop robot — would turn the Apple home stack into a persistent assistant surface, not just a speaker. If you build consumer services, assume Siri becomes a higher-intent, always-present broker for media, commerce, and device control inside Apple households.
Applied AIApple Cuts Jobs as It Reprioritizes Around AI
Apple trimming roles across Vision Pro, Siri, and Intelligent Systems while “reshaping” Siri around new AI infra is a clear signal that legacy voice stacks are being retired in favor of LLM-centric architectures. If you’re in Apple’s ecosystem, expect APIs, behaviors, and integration points around Siri and on-device intelligence to change materially over the next 6–18 months.