Chat modes - OverControl

OverControl Chat gives you three modes: Fast, Expert, and Vision. Choose a mode from the selector in Chat before sending your message. No API key or model configuration is needed.

Mode	Best for	What to expect
Fast	Writing, summaries, explanations, and everyday questions	Direct responses without an extended reasoning step. This is the default mode.
Expert	Complex analysis, coding, planning, and multi-step problems	More reasoning before the final answer. Responses can take longer and use more of your allowance.
Vision	Screenshots, photographs, diagrams, and other images	Answers that use the visual content of supported image attachments.

Fast

Use Fast when you want to move quickly from a question to a useful result. It works well for rewriting messages, summarizing text, brainstorming, explaining concepts, and straightforward coding help.

Turn these notes into a concise project update.
Include progress, blockers, and next steps. Keep the tone professional.

[Paste your notes here]

Fast uses Lume 3.5 Max without extended reasoning. It can also work with supported text documents and PDFs after Chat extracts their text.

Expert

Use Expert when a task has several steps, competing constraints, or details that need careful comparison. Give it the relevant context and ask for the reasoning or tradeoffs you need in the answer.

Review this implementation against the requirements below.
Identify correctness issues, explain their impact, and suggest the smallest useful fixes.

[Paste the requirements and code here]

Expert uses Lume 3.5 Max with deeper reasoning enabled. It can take longer than Fast. A more detailed reasoning process does not guarantee a correct answer, so check the result against your requirements.

Vision

Use Vision when the information you need is visible in an image. Attach a supported JPEG, PNG, or WebP image, wait for the upload to finish, then send your question.

Explain the error shown in this screenshot.
Identify the relevant text and suggest what I should check first.

Vision uses Lume Omni. It can describe visual content, read legible text, and help interpret diagrams. Small text, blurry images, and precise measurements can be difficult to interpret. Crop to the relevant area and provide a clear image. PDFs are processed as extracted text in Chat. Selecting Vision does not turn a scanned PDF into an image-based document analysis workflow. For visual pages, attach supported images of the relevant pages instead.

Choose a mode for attachments

Use Fast for a short document summary, Expert for a detailed comparison or review, and Vision for image content. Attachments are limited to 3 files per message, 5 MiB per file, and 10 MiB combined. See Files and images for supported formats and extraction limits. Chat’s attachment support is separate from the API’s multimodal capabilities; the mode selector does not enable audio uploads, web search, or code execution.

Get a more useful answer

State the task, provide the relevant context, and ask for the format you want. Keep follow-ups in the same conversation when they depend on earlier messages. Start a new chat for an unrelated task. Chat uses your OverControl account allowance. Review usage in the dashboard. Conversations are saved with a 30-day retention period; see Privacy and security before relying on Chat to keep important work.

​Fast

​Expert

​Vision

​Choose a mode for attachments

​Get a more useful answer

Fast

Expert

Vision

Choose a mode for attachments

Get a more useful answer