Google Colab CLI and Agent Skills: Cloud GPUs From Your Terminal

Melvin Vivas · X video post · 2026-06-09 · Open on X

Topics: AI Dev Tools & Productivity, Fine-tuning & Model Customization, AI Agents, Tool Use & MCP · Level: intermediate

Summary

Google released a Colab CLI and agent Skills that give you full Colab runtimes from the terminal. You can request GPUs or TPUs, run scripts remotely and open an interactive console. A built-in agent skill lets a coding agent do jobs like fine-tuning Gemma 3 1B on a dataset.

Key points

• Request GPU/TPU hardware from the terminal, e.g. colab --gpu A100.
• Run scripts remotely with colab exec.
• Interactive console/REPL access to Colab runtimes.
• Comes with a built-in agent skill, so coding agents can drive Colab for you.
• Example task: tell your agent "fine-tune Gemma 3 1B on this dataset".

Resources mentioned

• Google Colab CLI · tool · github.com · free
  A command-line tool for creating Colab GPU/TPU runtimes and running code on them from your terminal.
• Google Colab · tool · colab.research.google.com · free · recommended by both Bashiri Smith & Melvin Vivas
  Free GPU notebooks.
  Also in: Run Notebooks on a Free GPU with Google Colab (T4, 15GB VRAM) (Melvin Vivas on X · notes), Run Local Models on a Free GPU with Google Colab (T4) (Melvin Vivas on X · notes), Free Colab Notebooks for Local Model Fine-Tuning and Inference (Melvin Vivas on X · notes), Intent classification for support using GLiNER2.5-Decide notebook (Melvin Vivas on X · notes) and 12 more
• Gemma 3 1B · tool · huggingface.co · free
  A small open Gemma 3 model that is practical to fine-tune.

Try this

• Try provisioning a Colab GPU from the terminal with `colab --gpu A100` and running a script with `colab exec`.
• Have a coding agent fine-tune Gemma 3 1B on your own dataset through the Colab CLI skill.