A TPU, like a GPU, runs whatever model is loaded onto it, which means the hardware makes time-consuming runtime decisions as it interacts with each one. Frozen v2 would have some of those decisions for Gemini fixed in the transistors, reducing the number of steps the chip takes and the amount of data it shuttles around per query. That could cut response latency enough to enable new applications, one of the sources said.
The original Frozen design, spearheaded by Google DeepMind chief scientist Jeff Dean, went further and would've baked Gemini's weights themselves into the chip. Google set that proposal aside because silicon tied to a single model version would have too short a life cycle, according to the report. Frozen v2 freezes the architecture instead and leaves the weights updatable, so the chip stays useful across Gemini releases, but only for as long as Google builds them on the same underlying architecture. How much of the model to lock in is reportedly still undecided.
Google doesn't plan to produce Frozen v2 at TPU volumes and views this generation partly as a trial run for more specialized silicon as model designs settle. The chip would sit alongside, rather than replace, a TPU line that split into separate training and inference variants with the eighth generation announced at Cloud Next in April. Its 2028 target also comes in the same year Google has reportedly booked Intel to package more than 3 million TPUs .
Model-hardwired inference silicon already exists in demonstrations. Taalas, a Toronto startup that has raised more than $200 million from investors including Quiet Capital and Fidelity, launched its HC1 chip in February with Llama 3.1 8B permanently wired into an 815mm-squared die built on TSMC's N6 process. The company claims 17,000 tokens per second per user with no HBM on the package. Nvidia, meanwhile, struck a $20 billion deal in December to license technology from inference chip designer Groq.
Inside Google's TPU V8 strategy, delivering two chips for two crucial tasks at incredible scale
Google reportedly books Intel for packaging more than 3 million TPUs in 2028
Key considerations
- Investor positioning can change fast
- Volatility remains possible near catalysts
- Macro rates and liquidity can dominate flows
Reference reading
- https://www.tomshardware.com/tech-industry/SPONSORED_LINK_URL
- https://www.tomshardware.com/tech-industry/google-reportedly-developing-frozen-v2-chip-with-geminis-architecture-etched-into-the-silicon#main
- https://www.tomshardware.com/subscription
- TSMC eyes price hikes of up to 25% on chip production services in 2027, report claims — plans to raise baseline prices by 5% to 10% on advanced nodes
- Nvidia details Rubin architectural optimizations for inference – improvements target better performance and efficiency from the GPU to the rack
- Get 16GB of DDR5 RAM for $12 in this Newegg AM5 combo — score AMD's new Ryzen 7700X3D, 16GB of RAM, and Asus B850 motherboard for only $491
- 1994 Sega 32X gets first-person Backrooms game with raycasting engine — retro hardware tour de force includes yellow rooms, buzzing fluorescents, endless proced
- Russian drones spotted using screwed-on magnetic compasses as navigation aids — the on-board camera can occasionally tilt down to check bearings if satellite co
Informational only. No financial advice. Do your own research.