Brendan McKeag

Runpod’s new REST API lets you manage GPU workloads programmatically—launch, scale, and monitor pods without ever touching the dashboard.

We’ve upgraded Runpod CPU pods with Docker runtime and network volume support—giving you more flexibility, better storage options, and smoother dev workflows.

DeepSeek R1 remains one of the top open-source models. This post shows how you can run it efficiently on just 480GB of VRAM without sacrificing performance.

This follow-up to our “Hello World” tutorial walks through streaming output from a Runpod Serverless endpoint using WebSocket and base64 files.

New to serverless? This guide shows you how to deploy a basic "Hello World" API on Runpod Serverless using Docker—perfect for beginners testing their first worker.

Mistral Small 3 skips synthetic data entirely and still delivers strong performance. Here’s why that decision matters, and what it tells us about future model development.

Fine-tuning large language models can require hours or days of runtime. This guide walks through how to choose the right GPU spec for cost and performance.