Google 推出全新 Colab CLI，讓開發者與 AI 代理從終端機執行遠端 Colab GPU 與 TPU Python 程式

This week, Google AI team released the Colab CLI. The tool connects your local terminal to remote Colab runtimes. It lets developers and AI agents run code on cloud GPUs and TPUs. You stay in your terminal the entire time. The CLI is open source under the Apache 2.0 license. What is Google Colab CLI The Colab CLI is a command-line interface for Google Colab. You can create sessions, run code, and manage files from the terminal. Any agent with terminal access can call the tool. That includes Claude Code, Codex, and Google’s Antigravity. Google ships a prepackaged skill file named COLAB_SKILL.md. It gives agents built-in context on how to use the CLI. Installation uses a single uv tool install command from the GitHub repository. Copy CodeCopiedUse a different Browseruv tool install git+https://github.com/googlecolab/google-colab-cli A minimal session looks like this: Copy CodeCopiedUse a different Browsercolab new # provision a CPU session echo "print('hello')" | colab exec # run code colab stop # release the VM How the Commands Work The CLI groups commands into sessions, execution, files, and automation. colab new provisions a session, with CPU as the default. Add --gpu T4, --gpu L4, --gpu A100, or --gpu H100 for a GPU. TPU options are v5e1 and v6e1. colab exec runs Python from stdin, a .py file, or a notebook. The exec reads files locally and ships their contents. Local edits therefore need no separate upload step. colab stop terminates the session and releases the VM. Other commands cover files and authentication. colab upload and colab download move files between local and remote. colab drivemount mounts Google Drive, defaulting to /content/drive. colab auth authenticates the VM for Google Cloud services. colab exec and Artifact Recovery: The Core Loop The core loop is short. You provision a runtime, run a script, then pull results back. colab download retrieves models, datasets, and other files. colab log exports session history as .ipynb, .md, .txt, or .jsonl. So a remote run becomes a replayable notebook on your disk. colab repl and colab console give interactive access to the VM. colab install adds packages with uv, falling back to pip. Session metadata is stored at ~/.config/colab-cli/sessions.json. Example: Fine-Tuning Gemma 3 1B Google’s official release demonstrates an agent-driven fine-tuning job. The task fine-tunes google/gemma-3-1b-it using QLoRA. It trains on a Text-to-SQL dataset to improve SQL generation. The Antigravity agent runs the full pipeline with five commands. Copy CodeCopiedUse a different Browsercolab new --gpu T4 colab install transformers datasets peft trl bitsandbytes accelerate colab exec -f finetune_run.py colab log --output gemma_finetune_log.ipynb colab stop The agent then downloads the adapter model, adapter config, tokenizer config, and tokenizer. You can load and serve the fine-tuned model locally. No manual cloud provisioning command was typed by the user. Use Cases Offload laptop-bound training to a remote GPU or TPU without leaving the terminal. Let agents like Claude Code, Codex, or Antigravity run end-to-end ML pipelines. Fine-tune small models, such as Gemma 3 1B, with QLoRA remotely. Script notebook execution and export replayable .ipynb logs for reproducibility. Debug interactively on the VM through colab repl or colab console. Colab CLI vs Browser-Based Colab The CLI does not replace the notebook UI. It targets scripted, automated, and agent-driven work instead. Here is how the two workflows compare across common tasks. DimensionBrowser-Based ColabColab CLIInterfaceWeb notebook UILocal terminalAccelerator selectionRuntime menu in the browser--gpu / --tpu flags on colab newAgent useManual, UI-drivenAny terminal agent via commandsRun local scriptsPaste or upload into cellscolab exec -f script.pyArtifact retrievalManual download or Drivecolab download, colab logPackage install!pip inside a cellcolab install (uv, then pip)Session controlBrowser-managed runtimecolab new, colab stop, colab statusAgent skill fileNoneBundled COLAB_SKILL.md Strengths and Considerations Strengths: Terminal-native workflow fits scripts, CI, and agent loops. One command provisions T4, L4, A100, or H100 GPUs. exec ships local file contents, so no upload step is needed. Logs export to replayable notebook formats for reproducibility. Open source under Apache 2.0, with a bundled agent skill file. Works with multiple agents, not a single vendor’s tool. Considerations: Access requires authentication; the default strategy is oauth2. repl and console need a TTY when run interactively. Pipe stdin to use those two commands inside scripts. Compute still runs on Colab’s backend and its runtime model. Key Takeaways Google’s Colab CLI runs code on remote Colab GPUs and TPUs from your local terminal. One command provisions accelerators: colab new --gpu T4 through A100 and H100, plus TPUs. colab exec ships local .py and .ipynb files to the runtime without an upload step. Any terminal agent — Claude Code, Codex, Antigravity — can drive it via a bundled COLAB_SKILL.md. It is open source under Apache 2.0, and colab log exports replayable notebook logs. Marktechpost Visual Explainer #mtp-colab-cli-slider *{box-sizing:border-box;margin:0;padding:0} #mtp-colab-cli-slider{ background:#000!important;color:#f4f4f4!important; border:1px solid #2a2a2a!important;border-radius:14px!important; font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,Helvetica,Arial,sans-serif!important; max-width:820px;margin:24px auto;padding:0;overflow:hidden;line-height:1.5; -webkit-font-smoothing:antialiased } #mtp-colab-cli-slider hr,#mtp-colab-cli-slider p:empty,#mtp-colab-cli-slider del,#mtp-colab-cli-slider s{display:none!important} #mtp-colab-cli-slider .mtp-head{ display:flex;align-items:center;justify-content:space-between; padding:16px 26px;border-bottom:1px solid #1f1f1f!important } #mtp-colab-cli-slider .mtp-brand{font-size:13px;letter-spacing:.16em;color:#9a9a9a!important;text-transform:uppercase;font-weight:600} #mtp-colab-cli-slider .mtp-counter{font-size:13px;color:#ffffff!important;font-variant-numeric:tabular-nums;border:1px solid #333!important;border-radius:999px;padding:3px 11px} #mtp-colab-cli-slider .mtp-bar{height:1px!important;background:#1c1c1c!important;width:100%} #mtp-colab-cli-slider .mtp-bar span{display:block;height:1px!important;background:#ffffff!important;width:12.5%;transition:width .3s ease} #mtp-colab-cli-slider .mtp-stage{padding:30px 30px 6px 30px;min-height:330px;position:relative} #mtp-colab-cli-slider .mtp-slide{display:none!important} #mtp-colab-cli-slider .mtp-slide.active{display:block!important;animation:mtpfade .28s ease} @keyframes mtpfade{from{opacity:0;transform:translateY(6px)}to{opacity:1;transform:none}} #mtp-colab-cli-slider .mtp-eyebrow{font-size:12px;letter-spacing:.18em;text-transform:uppercase;color:#7d7d7d!important;font-weight:700;margin-bottom:12px} #mtp-colab-cli-slider h2{font-size:27px;color:#ffffff!important;font-weight:700;letter-spacing:-.01em;margin-bottom:10px;line-height:1.2} #mtp-colab-cli-slider h3{font-size:22px;color:#ffffff!important;font-weight:700;letter-spacing:-.01em;margin-bottom:14px} #mtp-colab-cli-slider .mtp-lead{font-size:17px;color:#cfcfcf!important;margin-bottom:14px;max-width:620px} #mtp-colab-cli-slider .mtp-meta{font-size:13px;color:#7d7d7d!important;margin-top:6px} #mtp-colab-cli-slider ul{list-style:none;margin:6px 0} #mtp-colab-cli-slider li{ font-size:16px;color:#dcdcdc!important;padding:9px 0 9px 22px;position:relative; border-bottom:1px solid #141414!important } #mtp-colab-cli-slider li:last-child{border-bottom:0!important} #mtp-colab-cli-slider li:before{content:"";position:absolute;left:2px;top:17px;width:6px;height:6px;background:#ffffff!important;border-radius:50%} #mtp-colab-cli-slider pre{ background:#0c0c0c!important;color:#e8e8e8!important;border:1px solid #262626!important; border-left:3px solid #ffffff!important;border-radius:8px!important; padding:15px 17px!impor