TPU 8i is co-designed with our Gemini research team to support low latency inference. Among the attributes that support this are large amounts of on-chip SRAM, enabling more computations to be done on chip without having to go to HBM for weights or KVCache state as often. The
HARDWARE
-
TPU 8t: 3X FP4 Performance Boost Over Ironwood
By
–
First, let's talk about TPU 8t, which is designed for large-scale training and inference throughput. The pod size is increased slightly to 9600 chips, and provides ~3X the FP4 performance per pod vs. Ironwood (8t has 121 exaflops/pod vs. 42.5 exaflops/pod for Ironwood). In
-
Google announces eighth-generation TPU chips for agentic era
By
–
I had a good time discussing yesterday's Google TPU v8t and v8i announcement at Cloud Next with Amin Vahdat along with @AcquiredFM hosts @gilbert and @djrosent
. The blog post announcement has lots of details about these new chips: https://
blog.google/innovation-and
-ai/infrastructure-and-cloud/google-cloud/eighth-generation-tpu-agentic-era/
… Here's a thread of -
CPUs Critical Infrastructure for Agentic AI Performance
By
–
🔥 Hot take: CPUs don’t get enough credit in agentic AI.
— SambaNova (@SambaNovaAI) 23 avril 2026
They prep data, route requests, and coordinate with accelerators, while also handling everything outside the model like code execution, DB queries, and validation.
Without them, inference can’t keep up 🦾
Do you agree… pic.twitter.com/RzBjFUHACdHot take: CPUs don’t get enough credit in agentic AI. They prep data, route requests, and coordinate with accelerators, while also handling everything outside the model like code execution, DB queries, and validation. Without them, inference can’t keep up Do you agree
-
kUPS GPU Optimization Achieves 49x Throughput Over RASPA
By
–
We’ve optimized kUPS specifically for GPU in collaboration with @nvidia , achieving up to 49× throughput over widely used software like RASPA for specific simulations.
-
kUPS: Molecular Simulation Engine for AI Workflows
By
–
Today at @iclr_conf 2026, I was excited to announce kUPS: a molecular simulation engine built for the AI era, optimized for GPU in collaboration with NVIDIA. kUPS is a plug-and-play, Python-native toolkit designed to integrate seamlessly with modern ML workflows.
-
AI Helps Build Apps for E-Readers and Flipper Zero
By
–
So it can basically understand how to build apps and tools for anything you throw at it. For example, I recently bought this device, a small e-reader. It fully understood the firmware, and now I am making apps for it! Or even making full apps/games for my Flipper Zero.
-
Pareto Frontiers: Extended Context and Improved Inference Speed
By
–
looks like new Pareto frontiers across everything:
— swyx 🇸🇬 (@swyx) 23 avril 2026
– Context: 400K context in Codex and a 1M in API
– API Pricing: $5/m input and $30/m output tokens.
– Codex improved its own inference speed 20% lol
– First generation co-designed with GB200 and GB300 NVL72
– 82.7% on… https://t.co/J5CL5fKmsq pic.twitter.com/nJG1vubSdNlooks like new Pareto frontiers across everything: – Context: 400K context in Codex and a 1M in API
– API Pricing: $5/m input and $30/m output tokens. – Codex improved its own inference speed 20% lol – First generation co-designed with GB200 and GB300 NVL72 – 82.7% on -
AI Model Creates and Deploys Flipper Zero Apps via USB
By
–
In this example, it created apps for my Flipper Zero through a USB connection and pushed them successfully to the device.
— Pietro Schirano (@skirano) 23 avril 2026
Just an idea, a cable, and a model that could actually make it real. pic.twitter.com/uWCsr0ydElIn this example, it created apps for my Flipper Zero through a USB connection and pushed them successfully to the device. Just an idea, a cable, and a model that could actually make it real.
-
GPT-5.5 Unlocks Vibe Hardware Era for AI Workflows
By
–
GPT-5.5 is the highest leverage tool I have ever touched.
— Pietro Schirano (@skirano) 23 avril 2026
For the first time, I don’t feel limited by what a model can do. I feel limited only by what I can imagine.
Training workflows. Impossible optimizations. Hardware experiments over USB.
The vibe hardware era begins. https://t.co/0mUw3YoWySGPT-5.5 is the highest leverage tool I have ever touched. For the first time, I don’t feel limited by what a model can do. I feel limited only by what I can imagine. Training workflows. Impossible optimizations. Hardware experiments over USB. The vibe hardware era begins.