MaxAI

Extracts full webpage DOM content and generates concise summaries using user-selected AI models (GPT-4o, Claude-3.5-Sonnet, Gemini-1.5-Pro, or Llama variants) via MaxAI's backend API. The extension captures page structure, filters boilerplate (navigation, ads), and sends cleaned content to the chosen model, returning a summary that preserves key information hierarchy. Supports both quick summaries (fast tier models) and detailed analysis (smart/reasoning tier models).

Enables interactive Q&A with PDF documents by extracting text content (via client-side or server-side OCR for scanned PDFs) and maintaining a conversation context with the selected AI model. Users upload PDFs or the extension detects PDFs on the current page, chunks the content, and sends queries alongside the document context to the backend API. Supports follow-up questions that reference previous answers, maintaining conversation state across multiple exchanges.

Performs web searches directly from the MaxAI sidebar and synthesizes AI-powered answers from search results. The extension queries the web, retrieves credible sources, and uses the selected AI model to synthesize a comprehensive answer with source citations and links. Supports follow-up questions and maintains search context across multiple queries. Answers include direct links to source pages for verification.

Generates images from text prompts using multiple generative models (DALL·E 3, FLUX.1 variants: schnell/dev/pro, Stable Diffusion 3 Medium) accessible from the MaxAI sidebar. Users enter a text description, select the desired model and quality settings, and the extension sends the prompt to the backend API for image generation. Generated images can be downloaded, edited, or used directly in webpages. Available on Pro and Elite tiers with unlimited generation on Elite.

Allows users to create, save, and manage custom AI prompts that can be reused across all MaxAI features. Users define prompt templates with variables (e.g., {{topic}}, {{tone}}), save them to a personal library, and apply them to selected text, images, or web content. The extension stores prompts locally or in the user's MaxAI account (depending on tier) and provides quick access via a dropdown menu. Supports prompt versioning and sharing (on higher tiers).

Maintains a persistent floating sidebar on every webpage that provides access to all MaxAI features (summarization, rewriting, search, chat, image generation) without page navigation. The sidebar preserves context across multiple webpages and tabs, allowing users to reference previous conversations or summaries while browsing. The sidebar can be minimized, repositioned, or hidden, and maintains state across browser sessions.

Integrates with the browser's right-click context menu to provide one-click access to MaxAI features for selected text or images. Users right-click on text, images, or links and select from a submenu of actions (summarize, rewrite, translate, explain, search, generate image, etc.). The extension captures the selected content and applies the chosen action using the user's default model and settings. Supports custom context menu items for frequently-used prompts.

Allows users to select from 5 AI model families (OpenAI, Anthropic, Google, Meta, Mistral) with 3 performance tiers (fast, smart, reasoning) for each action. The extension displays available models, their capabilities, and current quota usage before executing requests. Users can switch models per-request or set a default model. The extension tracks quota consumption across models and tiers, warning users when approaching limits. Quota resets monthly based on subscription tier.

Detects YouTube videos on the current page, extracts video transcripts (or generates them if unavailable), and provides one-click summarization or interactive Q&A about video content. The extension sends transcript text to the selected AI model, which generates summaries or answers questions about specific timestamps, topics, or speakers mentioned in the video. Supports both fast summaries and detailed analysis depending on model tier.

Integrates with Gmail and Outlook via browser extension hooks to detect email composition contexts and generate contextually appropriate replies. When a user opens an email, MaxAI analyzes the sender's tone, subject matter, and conversation history, then generates draft replies using the selected AI model. Users can customize tone (professional, casual, friendly, formal) and length before sending. The extension accesses email content through the browser DOM, not direct API access to email providers.

Detects comments and messages on LinkedIn, Facebook, YouTube, X/Twitter, and Instagram, then generates contextually appropriate replies using the selected AI model. The extension analyzes the comment/message content, platform context (public vs. private), and user's typical response style, then generates draft replies that can be customized before posting. Supports tone adjustment and multiple reply options for each comment.

Integrates with WhatsApp, Slack, Discord, Messenger, and Telegram to detect incoming messages and generate contextually appropriate replies. The extension analyzes message content, conversation history, and platform context (group chat vs. direct message), then generates draft replies using the selected AI model. Users can customize tone and length before sending. Integration works through browser-based versions of these platforms (WhatsApp Web, Slack Web, Discord Web, etc.).

Analyzes selected text on any webpage and generates rewritten versions with user-specified modifications (improve clarity, proofread, paraphrase, change tone, simplify language). The extension sends selected text and rewriting instructions to the chosen AI model, which returns multiple rewrite options. Users can preview each option and select the best fit before replacing the original text. Supports both inline editing and sidebar-based rewriting.

Translates webpage content into user-selected language using the chosen AI model, with options for full-page translation or bilingual side-by-side view. The extension extracts page text, sends it to the backend API for translation, and either replaces the original content or displays translated text alongside the original. Supports 100+ languages and maintains page structure and formatting during translation.

Analyzes selected text or concepts on a webpage and generates clear, accessible explanations using the chosen AI model. The extension sends selected text to the backend API with an instruction to explain in simple terms, and returns explanations that break down complex concepts into understandable components. Supports multiple explanation styles (beginner-friendly, technical, analogies) and follow-up questions for deeper understanding.

Analyzes images and screenshots on webpages or uploaded by users, extracting text via OCR and providing visual analysis using the chosen AI model. For images containing math problems, the extension recognizes the problem, solves it step-by-step, and explains the solution. Supports handwritten and printed text, diagrams, charts, and mathematical expressions. Users can upload screenshots or select images on the current page.