OpenAI says its Jalapeño chip delivered 1.5x-1.9x more AI work per watt
OpenAI says its Jalapeño chip delivered 1.5x-1.9x more AI work per watt and 1.7x-3.6x lower latency vs. Nvidia chips across GPT-OSS, DeepSeek R1, Kimi K2.5 1T
TL;DROpenAI's Jalapeño chip beats Nvidia on efficiency and speed in lab tests.
Why it matters: If real-world performance matches claims, it threatens Nvidia's AI inference dominance and could reshape GPU market dynamics.
Apple debuts its ‘most powerful chip ever’ in M5 Ultra and M6
TL;DRApple released faster M5 Ultra and M6 chips for its pro desktops.
Waymo plans to offer ride-hailing services in Germany, starting
TL;DRWaymo is bringing self-driving taxis to Munich by end of 2027.
Latest
Perplexity launches Portable Computer, a local AI agent platform running
Perplexity's new local AI runs on your own hardware with no cloud costs or token fees.
OpenAI’s Jalapeño chip is built for fast inference at scale, benchmarks
OpenAI's new chip handles AI inference faster and more efficiently than competitors.
The Supreme Court’s Mail-In Ballot Ruling Is a Step Toward Chaos in
Supreme Court clears path for stricter voting restrictions ahead of midterms.
Apple Mac Mini Revamp; Dick's Sporting Goods Falls on Outlook | Stock
Apple refreshes Mac Mini; Dick's Sporting Goods cuts outlook after Foot Locker struggles.
A detailed look at Jalapeño, OpenAI's ASIC developed with Broadcom
OpenAI built a custom AI chip with Broadcom that outperforms competitors on cost and efficiency.
Best RAM deals 2026 — discounts on DDR5 and DDR4 to beat the memory
RAM prices are dropping—good time to buy DDR5 or DDR4 memory.
Hot Chips 2026: Intel dives deep on Crescent Island AI accelerator —
Intel's new AI chip uses better cooling and memory for faster, efficient data center inference.
Uber launches optional safety feature allowing parents or guardians
Uber now lets parents watch live video of their teen's rides from the driver's phone camera.
Extremely Fast and Accurate Transcription with Granite Speech 5.0 Turbo
IBM released a faster, more accurate speech-to-text model called Granite Speech 5.0 Turbo.