✓ Free · Private · No sign-up
QuickAsk adds an instant chat to every website. Right-click selected text to ask about it, or open the chat widget for follow-up questions. Works with Google Gemini, Claude, OpenAI, Groq, DeepInfra, Cerebras, Ollama, and any OpenAI-compatible API.
One-time setup. After that, ask anything on any page.
Open ⚙ Settings and choose from the dropdown: Gemini, Claude, OpenAI, Groq, DeepInfra, Cerebras, Ollama, or any custom OpenAI-compatible endpoint.
Each provider's link is right there in Settings. Get a free key (Gemini and Groq have generous free tiers) and paste it in. Done.
Right-click any selected text → Ask AI, or click the 💬 button in the bottom-right of any page
for a chat.
One extension, one chat UI. Your choice of model and price tier.
Designed to stay out of your way until you need it.
Select text on any page, right-click → Ask AI about: "..." for an instant one-shot answer.
Click the 💬 button (bottom-right) for a multi-turn chat with full markdown rendering.
Save any conversation as a .txt file or a printable PDF. Never lose a good answer.
If a request fails, your question stays on screen so you can retry, edit, and resend without retyping.
First-class support for reasoning models: OpenAI o-series, Claude extended thinking, DeepSeek-R1, Gemini 3.
One-click theme toggle in the popup. Matches your eyes, not your OS.
No servers. No analytics. Your API key and chats live in your browser only. Requests go straight to the AI provider.
Set a custom system prompt to shape the AI's tone, length, or persona across every chat.
No subscription, no ads. You pay your AI provider, never us. Tips are appreciated but never required.
Other questions? Email us — we read every message.
The extension itself is 100% free and ad-free. You pay only your AI provider directly (Gemini and Groq have generous free tiers, so you can use QuickAsk without paying anything).
No. There is no QuickAsk server. Your API key, conversations and settings
live in chrome.storage, your browser's own storage. When you ask a question,
the request goes directly from your browser to the AI provider you picked.
The developer of this extension never sees your key, your prompts or the answers.
See the Privacy Policy for full details.
If you want to try it free, get a Google AI Studio key. It's free and supports the latest
gemini-3-flash-preview.
If you want speed, try Groq or Cerebras (both free tiers, blazing fast).
For best quality, use Anthropic Claude (claude-opus-4-7) or
OpenAI (gpt-5.5).
Fully supported. QuickAsk auto-detects models that use extended thinking
(OpenAI o1/o3/o4 & GPT-5 reasoning variants,
Claude opus-4-7 / sonnet-4-6, DeepSeek-R1 / V4,
Gemini 3, Qwen QwQ) and adjusts the request format accordingly (using
max_completion_tokens, enabling Claude's thinking blocks, etc.).
Reasoning output is hidden. You get the clean final answer.
Yes. Open ⚙ Settings → Show advanced → System prompt. Examples worth trying:
• "Be concise and scientific. Use bullet points and cite sources when possible."
• "Reply in 2-3 sentences max. Plain English, no jargon."
• "You are a senior code reviewer. Spot bugs first, then suggest improvements."
• "Always answer in German. Use a friendly, casual tone."
• "Format every answer as Markdown with a TL;DR at the top."
Yes. Pick Ollama (local) from the provider dropdown. The default endpoint
is http://localhost:11434, no API key required. Make sure Ollama is running and
the model you choose is pulled (ollama pull llama3.2).
Google Chrome 88+, Microsoft Edge (Chromium), Brave, Arc, and other Chromium-based browsers.
Add QuickAsk to Chrome, pick a provider, paste your key, and ask anything on any page.
QuickAsk is free and ad-free. If it saves you a few searches a day, a coffee keeps the project alive and updated for everyone.
QuickAsk is an independent, community-built extension. It is not affiliated with, endorsed by, or sponsored by Google, Anthropic, OpenAI, Groq, DeepInfra, Cerebras, OpenRouter, or Ollama. All trademarks belong to their respective owners.
By installing this extension you agree that the author accepts no liability for account issues, API usage charges, lost work, or any other outcome of using it. There is no warranty, express or implied. You are responsible for the API costs you incur with your chosen provider.