Google unveils TPU 8tx and TPU 8i Max for AI workloads
TPU 8tx, designed with AMD, targets confidential training and inference; TPU 8i Max doubles memory in one package.
Stories per day · 30 days
2 RSS
Relatedgoogle6inference4pytorch4torchtpu3ai-chips2confidential-computing2gemma2google-cloud2ai-hardware1amd1attention1gpu1hugging-face1huggingface1linear-attention1llm-d1nvidia1open-source1
15 stories, newest first · one card per story
Stories tagged #tpu
Systolic Linear Attention optimizes gated delta attention for Google TPUs
The method achieves 1.6x to 2.5x faster performance on TPU v7x and 11-24% faster training for MoE models compared to KDA.
Google's Project Suncatcher satellite runs Gemma 4 from orbit
Onboard TPUs completed a space haiku prompt; Google next plans thermal and radiation tests and twin satellites in 2027.
Google’s Suncatcher satellite is online and running Gemma 4
The prototype carries one TPU, generates up to 1 kW, and can power the chip for about twenty minutes.
1
Google Cloud ships llm-d 1.0 with cache-aware routing
The captured page is a cookie notice, so routing, TPU, and GPU claims are not confirmed in the body.
7
Google open-sources TorchTPU for running PyTorch on TPUs
TPU support also landed in vLLM, SGLang, TorchTitan, Ray, and Hugging Face libraries, with two pieces still in preview.
Google open-sources TorchTPU so PyTorch apps can run on TPUs
Amazon released TorchNeuron a day earlier to port PyTorch apps to Trainium without rewriting them.
Google open-sources TorchTPU, a native PyTorch back end for TPUs
The feed only shows a Hacker News thread with 42 points and 12 comments.
Hugging Face announces native TorchTPU support
The post title says Hugging Face now supports TPUs natively through TorchTPU; no further details were provided.
AI is now capable of developing its own inference hardware
A GitHub project, openTPU, claims AI can design its own inference hardware.
Google thinks SpaceX’s Starship has to launch 1,600 times before space data centers get off the ground
Google launched a TPU into orbit on SpaceX to test Project Suncatcher's space data-center concept.
E3J: An Efficient and Open-Source Backend for Euclidean Equivariant Operations on GPU and TPU
Open-source e3j speeds Euclidean-equivariant GPU and TPU kernels, beating cuEquivariance on a MACE simulation.
Google's first Suncatcher orbital data center test launches October 1
Google's first Suncatcher orbital data-center test launches October 1 to run Gemini on TPUs in space.
Google is sending an AI satellite into space next week
Google will launch a TPU satellite on a SpaceX Falcon 9 on October 1 to test AI compute in orbit.
3
MaxKernel: Agentic Kernel Generation for TPUs
Researchers open-source MaxKernel, a multi-agent LLM system that generates and optimizes TPU kernels matching expert hand-tuned baselines on JaxBench.