DEV Community

TPU Series' Articles

Back to xbill's Series
tpu-management: a Claude Code skill for running Gemma 4 on Cloud TPUs
xbill profile

tpu-management: a Claude Code skill for running Gemma 4 on Cloud TPUs

#googlecloud #tpu #vllm #claudecode
8
Comments
3 min read
Gemma 4 E2B on a Single TPU v6e Chip: A Serving Deep Dive
xbill profile

Gemma 4 E2B on a Single TPU v6e Chip: A Serving Deep Dive

#tpu #llm #vllm #googlecloud
10
Comments 1
8 min read
Serving Gemma 4 E2B on a TPU v6e-1: what Trillium buys, and what it doesn't
Cover image for Serving Gemma 4 E2B on a TPU v6e-1: what Trillium buys, and what it doesn't
xbill profile

Serving Gemma 4 E2B on a TPU v6e-1: what Trillium buys, and what it doesn't

#tpu #vllm #llm #gcp
2
Comments
20 min read
The unofficial TPU migration guide: Cloud TPU API to Compute Engine
xbill profile

The unofficial TPU migration guide: Cloud TPU API to Compute Engine

#tpu #gcp #vllm #devops
8
Comments 2
17 min read
Deploying a QAT Checkpoint Your Serving Stack Can't Load: Gemma 4 E2B in Pure JAX on One TPU
xbill profile

Deploying a QAT Checkpoint Your Serving Stack Can't Load: Gemma 4 E2B in Pure JAX on One TPU

#tpu #jax #llm #quantization
8
Comments
10 min read
Repacked QAT Gemma 4 on One TPU v5e: 12B Serves at 675 Tokens per Second
Cover image for Repacked QAT Gemma 4 on One TPU v5e: 12B Serves at 675 Tokens per Second
xbill profile

Repacked QAT Gemma 4 on One TPU v5e: 12B Serves at 675 Tokens per Second

#gemma #googlecloud #machinelearning #llm
8
Comments
9 min read