g5g vs g6 for LLM Serving: the Same Code, and 3.7x the Throughput
This article compares two AWS GPU instance families for serving a small language model, using a payload that is byte-identical on both. The older fa…
DevOps
⚑ Сообщить о проблемеПоследние новости DevOps от Tech News
Gemma 4 in Pure JAX: What Ports from TPU to GPU, and What Doesn't
This article is about running a hand-written Gemma 4 port in pure JAX on three different accelerators, and about the two places the abstraction leak…
Deploying a QAT Checkpoint Your Serving Stack Can't Load: Gemma 4 E2B in Pure JAX on One TPU
Cloud TPU v6e-1 ( ct6e-standard-1t , one v6e chip, 32 GB HBM), Compute Engine flex-start, europe-west4-a. All timings below measured 2026-08-19 unle…
One TPU Chip, Eight Agents: Serving Small Agent Workloads with Raw JAX
Cloud TPU v6e-1 ( ct6e-standard-1t , one v6e chip, 32 GB HBM), GCE flex-start, europe-west4-a. vLLM baseline measured 2026-07-21. The workload nobod…
PyTrees Are Not One Thing: JAX, PyTorch, and TensorFlow Compared
PyTrees look deceptively simple. You flatten a nested Python object into leaves, keep a structure descriptor, and later rebuild or map over the same…
You Can Now Train Quantum Signal Processing Phase Angles with Gradient Descent
A reproducible benchmark and JAX‑traceable implementation is now available – no analytic solver required. Quantum Signal Processing (QSP) is one of…