QuickAsk – Instant AI Chat

About

QuickAsk: instant AI chat, anywhere you browse.
 
Highlight a sentence on any page, right-click, and get an AI answer in seconds. Or click the floating chat bubble in the corner for a full multi-turn conversation, with markdown rendering, code highlighting, and conversation export.
 
No account. No sign-up. No telemetry. Just your API key and any AI provider you trust.
 
━━━━━━━━━━━━━━━━━━━━━━
✨ KEY FEATURES
━━━━━━━━━━━━━━━━━━━━━━
 
🖱 RIGHT-CLICK QUICK ASK
• Select any text on a page, right-click, and pick “Ask AI about: …”
• Selected text is sent as context alongside your question
• Modal closes after answering, so no clutter is left behind
 
💬 PERSISTENT CHAT WINDOW
• Click the floating 💬 button (bottom-right) for an ongoing multi-turn conversation
• Full markdown and code-block rendering with language labels
• Save chat as PDF or plain .txt
• Right-click anywhere on a page and choose “Open QuickAsk Chat” to launch it without selecting text
• Toggle the floating button off in Settings if you only want the right-click menu
 
⚙ WORKS WITH 9 PROVIDERS (and any OpenAI-compatible endpoint)
Each provider ships with one-click presets that fill in the endpoint and model for you:
• Google Gemini - fast, capable models with a free tier
• Anthropic Claude - Haiku, Sonnet, and Opus, with extended thinking support
• OpenAI - the GPT family
• Groq - open models with ultra-fast inference
• DeepInfra - affordable open-weight and reasoning models
• Cerebras - open models on extremely fast hardware
• OpenRouter - access 100+ models behind a single key
• Ollama  run local models fully offline, no key needed
• Custom - any service exposing an OpenAI-compatible /v1/chat/completions endpoint
 
Pick a provider, click a preset, and you’re ready in seconds. Advanced users can paste any custom URL and model name.
 
🧠 REASONING MODEL SUPPORT
• Automatically detects thinking-style reasoning models and adjusts behavior accordingly
• Enables extended thinking on models that support it
• Uses the right token limits (max_completion_tokens) and temperature settings per model class
 
📝 CUSTOM SYSTEM PROMPT
• Tell the AI to reply in a specific language, be concise, use bullet points, or anything else that fits your workflow
• Applied to every request automatically
• 8 one-click presets included: Concise & scientific, Short & plain, Translate to English, Code reviewer, TL;DR + Markdown, ELI12, German + casual, Summarize in bullets
 
🌗 LIGHT / DARK THEME
• Toggle between light and dark mode from the popup or settings
• Choice is remembered across sessions
 
🔒 PRIVACY FIRST: YOUR API KEY STAYS WITH YOU
• Your API key is stored only in your own browser (chrome.storage.sync)
• Requests go directly from your browser to the AI provider you configure
• No middleman server, no developer-controlled proxy
• No analytics, no tracking, no cookies, no telemetry
• Nothing is collected, logged, or transmitted to any third party
 
📄 MARKDOWN RENDERED
• Headings, bold, italic, bullet lists, numbered lists, blockquotes
• Inline code and fenced code blocks with language labels
• Links open in a new tab
 
💾 EXPORT CONVERSATIONS
• Download as .txt for archiving
• Print or save as PDF via the browser’s native print dialog
 
━━━━━━━━━━━━━━━━━━━━━━
🚀 HOW TO GET STARTED
━━━━━━━━━━━━━━━━━━━━━━
 
Install the extension
Click the QuickAsk icon in the toolbar
Pick a provider, then click a preset to fill the endpoint and model
Paste your API key
Click Save settings
Highlight any text on a page, right-click, and choose Ask AI. Done!
 
━━━━━━━━━━━━━━━━━━━━━━
🔑 WHERE TO GET AN API KEY
━━━━━━━━━━━━━━━━━━━━━━
 
Several supported providers offer a free tier or free credits to get started, and others are low-cost pay-as-you-go. Ollama runs fully locally on your machine and needs no key at all. Links to each provider’s key page are shown inside the extension’s settings when you select that provider.
 
━━━━━━━━━━━━━━━━━━━━━━
🌐 CUSTOM & OPEN-SOURCE COMPATIBLE
━━━━━━━━━━━━━━━━━━━━━━
 
QuickAsk works with any server that implements the OpenAI Chat Completions API format (POST /v1/chat/completions). This includes local servers like Ollama and LM Studio, self-hosted backends like vLLM, and any other OpenAI-compatible service. Just paste the base URL in the Endpoint field and enter your model name.
 
━━━━━━━━━━━━━━━━━━━━━━
💻 BROWSER SUPPORT
━━━━━━━━━━━━━━━━━━━━━━
 
✅ Google Chrome 88+
✅ Microsoft Edge (Chromium)
✅ Brave Browser
✅ Arc
✅ Other Chromium-based browsers
 
━━━━━━━━━━━━━━━━━━━━━━
📬 SUPPORT & FEEDBACK
━━━━━━━━━━━━━━━━━━━━━━
 
Love the extension? Leave a ⭐⭐⭐⭐⭐ review. It really helps!
Questions or issues? Email: [email protected]
Want to support development? Use the 💝 Support button in the popup, or visit https://buymeacoffee.com/science64
 
━━━━━━━━━━━━━━━━━━━━━━
⚠ DISCLAIMER
━━━━━━━━━━━━━━━━━━━━━━
 
QuickAsk is an independent, self-built browser extension. It is not affiliated with, endorsed by, or sponsored by any of the AI providers mentioned. All trademarks belong to their respective owners.
 
Use of this extension is subject to the terms of service of the AI provider you configure. The author accepts no liability for API costs, account issues, or any other outcomes arising from use of this extension.