Skip to Content

Connect CloudGPU to Your AI Agent

Draft a customer reply, turn meeting notes into an action plan, or extract information from an image. Ask your assistant to choose a CloudGPU model and shape the result for your next step.

2 toolsAPI key

20,000 free app calls every month(paid services billed separately)

Try a CloudGPU Task

Copy a prompt into your agent with the oo Skill installed. Add your page links and source notes before running it.

oo SkillCloudGPU

Turn meeting notes into next steps

Extract decisions, owners, and open questions from the notes you supply.

/oo CloudGPU: Use [model] to organize these meeting notes into decisions, owners, and next steps. Mark any missing owner or deadline as unconfirmed: [notes].

Draft a reply from your talking points

Give the model the message and the facts to include, then review the proposed reply.

/oo CloudGPU: Use [model] to draft a concise reply to this customer message. Confirm Wednesday delivery, acknowledge the delay, and do not offer a discount: [message].

Get a second model’s perspective

Ask two available models the same question and compare their assumptions and reasoning.

/oo CloudGPU: Send this proposal to [model A] and [model B]. Ask each to identify risks and unanswered questions, then compare where they agree and differ: [proposal].

Read the details in an image

Turn screenshots and images into useful summaries with a vision model.

/oo CloudGPU: Use a vision model to read this screenshot and organize the key information into a table: [image].

Turn free-form text into organized data

Extract names, dates, and key details into a consistent format.

/oo CloudGPU: Extract the company, contact, requested service, and deadline from these customer inquiries, with one row per inquiry: [inquiries].

Start Working with CloudGPU

  1. Install OOMOL

    Set up oo CLI and the oo Skill in your agent.

    Install OOMOL →
  2. Connect CloudGPU

    Create an API key in the CloudGPU platform and connect it.

    Connect CloudGPU→
  3. Give Your Agent a Task

    Use /oo with the pages to read, the changes to make, and where to save the result.

    Connection Guide →
  • A CloudGPU API key from https://cloudgpu.app/api/console.
  • Account credit for inference requests.
  • Generation uses your CloudGPU account credit. Available features depend on the model you choose.

CloudGPU Tools

Connect CloudGPU to work with multiple AI models in one place. Generate and refine content, continue conversations, analyze images with vision models, and return organized results for other tasks.

Choose from your available models

See which models your key can access before assigning a task.

Create, analyze, and organize

Write and refine content, analyze images, extract structured results, and request tool calls for your workflows with models that support these features.

2 tools

Create Chat Completion

Generate a CloudGPU chat completion through the OpenAI-compatible endpoint. This is a billed, non-streaming inference request.

cloudgpu.create_chat_completion

List Models

List the model IDs CloudGPU serves, with model metadata, without running inference.

cloudgpu.list_models

Frequently Asked Questions

How do I choose a model?

Ask your assistant to list the models available in your CloudGPU account, then choose one for writing, image analysis, or reasoning. You can also compare how different models handle the same task.

Does connecting run paid inference?

No. Connection validation only queries your account. Generation requests use account credit.

Will a drafted reply be sent automatically?

These actions generate text. Sending an email or saving a document requires a separate connected app and an explicit instruction.