Olly modes

When you chat with Olly, you can pick how it answers your prompts by selecting a mode before sending. Each mode is optimized for a different balance of speed, depth, and reasoning, and you can switch modes at any time during a conversation.

Olly supports three modes:

• Pro mode (default): a skill-based architecture optimized for both speed and quality
• Focus mode: deeper reasoning for complex investigations
• Fast mode: quick responses for simple questions

Shows the mode picker in the Olly chat input bar.

Pro mode (default)

Pro mode is Olly's default mode. It runs on a skill-based architecture: Olly applies a curated set of specialized skills to each request, the same pattern modern AI agents use to tackle complex tasks. The result is faster and higher-quality answers than Focus mode for most observability work.

• Models: pick from the same higher-tier list as Focus mode, plus Gemini 3.1 Pro — currently available in Pro mode only. See Model selection.
• Best for: most observability tasks — investigations, root-cause analysis, day-to-day questions.
• Faster and higher quality than Focus mode for the same prompt.

Pro mode is identified by a New tag in the mode picker.

Examples:

• "Why did latency spike after the last deployment?"
• "What's the current error rate for the checkout service?"
• "Compare today's API errors to last week."

Focus mode

Focus mode is optimized for deeper reasoning on complex investigations.

• Models: pick from a curated list of higher-tier GPT and Claude models. See Model selection.
• Best for: complex investigations, root-cause analysis, exploratory observability questions.
• Deeper analysis and reasoning.
• Takes longer to respond than Fast mode.
• Uses multiple specialized sub-agents, each acting as an expert in a specific domain (for example, logs agent).

Pick Focus mode when you want explicit multi-step reasoning with specialized sub-agents.

Examples:

• "Investigate the root cause of intermittent 5xx errors."
• "Correlate error logs with recent infrastructure changes."

Fast mode

Fast mode is designed for quick responses and lightweight tasks.

• Models: pick from a curated list of lower-tier GPT and Claude models. See Model selection.
• Best for: simple questions, quick lookups, basic data queries.
• Very fast response time with simplified reasoning optimized for speed.

Use Fast mode when you need a quick answer to a simple question, fast lookup, or basic data query where speed matters more than deep reasoning.

Examples:

• "Show error rate for checkout service in the last hour."
• "What is the current CPU usage of node-3?"
• "List alerts fired in the last 10 minutes."

When to use each mode

Next steps

Pick the AI model that best fits your speed, depth, and cost needs in Model selection.