The Runpod Blog

Part one of AI Infrastructure 101, a seven-part series.

The LTX-2.5 weights are out, with day-zero ComfyUI support. Here's what actually changed, and what you need to get generating on Runpod today.

A practical guide for accurately calculating the VRAM requirements for full-parameter model fine-tuning, explaining why standard inference-based rules of thumb are insufficient and offering equations to help users properly size their compute resources.

Customizability is the most underrated idea in AI right now. Runpod CEO Zhen Lu on why a model tuned on your data beats a bigger one on the job you actually have.

A practical guide to expanding multi-node GPU workloads in place.

A hands-on tutorial for wiring GPU-backed tools into an MCP server, and hosting the compute on Runpod Serverless.

Learn how Runpod's Model Store eliminates redundant downloads and uses a tiered architecture with smart scheduling to drastically reduce AI model cold start times.