In the news
Introducing vLLM TorchTPU
vLLM · Inferact and Google Teams · Published · 3 min read
In 30 seconds
- llm
- kernel
- serving
- kv cache
- vllm
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.