It is a sign of how fast things have changed that this is what OpenAI announced as the future of enterprise AI agents back in October 2025. They killed Agent Builder in June. (But a surprising number of enterprise products still use this model for agents).
BIG TECH
-
Boston Dynamics Integrates Gemini AI With Spot
By
–
Spot Gets Smarter with Gemini #AI
— Ronald van Loon (@Ronald_vanLoon) 25 août 2026
by @BostonDynamics#Robotics #Engineering #ArtificialIntelligence #Innovation #Technology pic.twitter.com/BTu0DYtl9HSpot Gets Smarter with Gemini #AI
by @BostonDynamics #Robotics #Engineering #ArtificialIntelligence #Innovation #Technology -
NVIDIA Ships Vera Rubin NVL72 Production Racks With Automated Assembly
By
–
NVIDIA Vera Rubin NVL72 production racks are here.
— NVIDIA (@nvidia) 25 août 2026
The compute tray is engineered for fast compute, assembly, and serviceability. Manufacturing is 100% automated, and every tray goes together in one minute.
Congratulations to @Microsoft on the first operational Vera Rubin NVL72… pic.twitter.com/Q1GyYjnD1mNVIDIA Vera Rubin NVL72 production racks are here. The compute tray is engineered for fast compute, assembly, and serviceability. Manufacturing is 100% automated, and every tray goes together in one minute. Congratulations to @Microsoft on the first operational Vera Rubin NVL72
-
Reports Describe OpenAI’s Large Bel Model and Anthropic Compute Constraints
By
–
OpenAI has reportedly (per @synthwavedd ) finished training “Bel,” a massive successor to “Doug” with over 10 trillion total parameters, probably the base for GPT-6. But more interesting: Anthropic reportedly lacks the compute to answer OpenAI’s upcoming Astra launch this year.
-
GPT-Astra Uses Codex to Optimize Jalapeño Kernels for Open Models
By
–
interesting detail in OpenAIs blogpost to "Jalapeño": GPT-Astra used Codex to write and optimize low-level kernels that brought three previously unplanned open-weight models to high performance on OpenAI’s Jalapeño chip within two months. For selected attention and
-
Commentator Sees Apple Challenging NVIDIA in Local AI
By
–
In other words, Apple might beat NVIDIA in Local AI if the later doesn’t change course
-
OpenAI Plans Jalapeño Deployment and Outlines Next-Generation Chip Roadmap
By
–
We plan to begin deploying Jalapeño in OpenAI’s compute infrastructure by year-end. It’s the first step in a multigenerational roadmap: Gen 2 is deep in development, and Gen 3 is taking shape. Each generation will push efficiency and speed further.
-
OpenAI Reports Higher Throughput and Lower Latency From Jalapeño
By
–
Since announcing Jalapeño, our first custom inference chip, we’ve been testing it and the system around it.
— OpenAI (@OpenAI) 25 août 2026
The results show a major advance: more intelligence from every watt and faster responses, delivering both higher throughput and lower latency in one architecture without… pic.twitter.com/vj7VOrA8pPSince announcing Jalapeño, our first custom inference chip, we’ve been testing it and the system around it. The results show a major advance: more intelligence from every watt and faster responses, delivering both higher throughput and lower latency in one architecture without
-
Gemini Enterprise Adds Legal and Finance Plugins Bundling Apps and Skills
By
–
Google announced its first plugins for Gemini Enterprise, tailored for legal and finance use cases.
— 🚨 AI News | TestingCatalog (@testingcatalog) 25 août 2026
These plugins combine selected Apps and Skills under purpose based bundle instructions.
It is also expected that bundles eventually will arrive to consumer Gemini. What worries… https://t.co/WHGAvTzIQA pic.twitter.com/QhlqFyygFwGoogle announced its first plugins for Gemini Enterprise, tailored for legal and finance use cases. These plugins combine selected Apps and Skills under purpose based bundle instructions. It is also expected that bundles eventually will arrive to consumer Gemini. What worries
-
OpenAI’s Jalapeño Benchmarks Show Efficiency and Latency Gains
By
–
OpenAI announced initial performance results of their own Jalapeño chip for AI inference.
— 🚨 AI News | TestingCatalog (@testingcatalog) 25 août 2026
> 1.5–1.9× more AI work per watt
> 1.7–3.6× lower end-to-end latency
> 2.1–4.1× higher performance on highly interactive workloads
Looking forward to seeing this deployed! https://t.co/x5HJPejq54 pic.twitter.com/CshpV23C0HOpenAI announced initial performance results of their own Jalapeño chip for AI inference. > 1.5–1.9× more AI work per watt
> 1.7–3.6× lower end-to-end latency
> 2.1–4.1× higher performance on highly interactive workloads Looking forward to seeing this deployed!