Video summary

Use Claude Code UNLIMITED (No GPU, No Limits!)

Main summary

Key takeaways

Technology

Summary of the video (tech concepts, setup steps, and how-to)

The video is a tutorial explaining how to use “Cloud code” for free on Windows by pairing it with Ollama to run AI models locally (or pull cloud-hosted models via Ollama).


1) Pricing limitation and the goal (“Cloud code free”)

  • The creator notes that Cloud code has a free version, but missing Cloud code features in that tier.
  • They describe a workaround: run models locally using Ollama, so Cloud code can use models without paid model hosting.

2) Install Ollama on Windows

  1. Open a browser and go to: ollama.com
  2. Install using one of these options:
    • Downloadable installer, or
    • A PowerShell command (the creator uses PowerShell)
  3. Verify installation in CMD by running a help command such as:
    • ollama -h

3) Install Cloud Code (Quick Start from official docs)

  1. Navigate to: code.cloud.com/docs
  2. Follow the Quick Start instructions.
  3. Choose the CMD-based installation (copy/paste command in CMD).
  4. Verify Cloud code launches successfully using a cloud-type command.

4) Create a local workspace for Cloud projects

  • Create a folder named (as shown) Cloud_projects (example: D:\Cloud_projects)
  • Open CMD inside that folder to run subsequent Cloud Code commands there.

5) Use AI agent models inside Cloud Code via Ollama

  • The creator mentions “launch” commands in Ollama for things like Cloud / Cloud Codex / Open Code, treated as AI agents.
  • They demonstrate starting a selected model/agent using an Ollama launch command.
  • Examples of models mentioned in the interface include names like “Gemini,” “quant,” “GLM,” and others.
  • They differentiate:
    • Local models: run locally without needing extremely powerful hardware.
    • Cloud-based models: referenced because the user’s computer can’t host the full model locally.

6) Pull a cloud-hosted model using Ollama (run “cloud models”)

  1. Search the Ollama library for a specific model (example used: GPT-OASIS).
  2. Copy the model reference for the cloud version (e.g., the reference that includes “cloud”).
  3. Use CMD with:
    • ollama pull <model-name>

Key detail: The creator claims Ollama pulls the manifest locally, while the actual model runs via the cloud (internet required). Therefore, the model cannot be used offline.


7) Launch Cloud Code and select the pulled cloud model

  • From within the project directory, run an Ollama/Cloud command to launch Cloud code, such as:
    • ollama launch Cloud
  • Select the model using arrow keys, then confirm.
  • On first run, Cloud code prompts for settings such as:
    • Theme
    • Sign in to Ollama

8) Example project: generate a Tic-Tac-Toe game

  • They prompt Cloud Code to create a tic-tac-toe game using:
    • HTML, CSS, JavaScript
  • The tool generates files such as:
    • index.html
    • style.css
    • JavaScript file(s)
  • They demonstrate gameplay:
    • X wins / O wins
    • “Game over” and “reset”

Main speakers/sources

  • Speaker: Unspecified individual host (no named person given)
  • Primary sources referenced:
    • Ollama website: ollama.com
    • Cloud Code documentation: code.cloud.com/docs (Quick Start)
  • Tools/platforms used:
    • Ollama (local models + cloud-model pulling via Ollama)
    • Cloud Code
    • Windows PowerShell and CMD

Rate this summary

Your feedback will help improve summaries.

Improve this summary

Reprocess with a stronger model when the summary feels incomplete or inaccurate.

Pro

Translate summary in another language

Pro

Ask questions to this video

Chat for follow-up questions, clarifications, and source-backed answers.

Coming soon

Share this summary

Original video