Every figure, date, quota and quoted description the finished picture puts on screen, chased to a primary source. Anything that could not be reached at a primary source is listed under Not checked at the end and is not drawn as a number.
The free Gemini plan is listed on Google’s own plan page as $0 with a Google Account, and includes chat, image generation and editing, Deep Research, Gemini Live, Canvas and Gems, with access to the current Flash model and varying access to the Pro model.
NotebookLM was renamed Gemini Notebook on 16 July 2026. Google’s announcement states it is “the same standalone product, now doing more across the Google ecosystem”, that notebooks become accessible inside the Gemini app with cross app syncing, and that Google plans to integrate notebooks into Search’s AI Mode.
The same post describes the code execution capability: Google has given “every notebook a secure cloud computer”, allowing Gemini Notebook to “write and execute code natively, helping you conduct complex data analysis grounded in your sources.”
AI Mode is documented as Google’s AI search experience with follow up questions, links to the web, and entry from an AI Overview. Google Lens transitions into a live AI Mode conversation via the Live option, and context carries forward across a session.
Google’s AI products page carries the line “Create interactive web components in Google AI Studio with a single prompt.”
Build mode, publishing and building an app from a drawing are documented by Google directly: AI Studio’s learn material includes “Build an App from a Single Mockup or Sketch”, and annotation mode lets you draw on the app preview and attach instructions to specific UI elements. Projects can be bootstrapped into a repository and deployed to Cloud Run.
Firebase Studio is sunsetting on 22 March 2027. New workspace creation and user signup were disabled on 22 June 2026. Google recommends migrating to Antigravity for a code first agentic experience, or to Google AI Studio for web based prototyping.
Google’s announcement post states the free individual tier offers “up to 180,000 code completions per month”, describes the capacity as “practically unlimited”, and is available in Visual Studio Code, JetBrains IDEs and GitHub. Gemini Code Assist for GitHub provides free AI powered code reviews for public and private repositories.
The comparison in the script is Google’s own framing: the post says other popular free coding assistants offer “only 2,000 code completions per month”, which Google describes as “90 times more code completions per month”. Published 25 February 2025.
The published free tier for logging in with a Google account through Gemini Code Assist for individuals is 1,000 model requests per user per day and 60 model requests per user per minute.
Google’s Jules usage limits list 15 daily tasks on a rolling 24 hours and 3 concurrent tasks on the base plan. Jules in Pro lists 100 daily and 15 concurrent; Jules in Ultra lists 300 daily and 60 concurrent.
Jules is documented as integrating with GitHub, running “in a virtual machine where it clones your code, installs dependencies, and modifies files”, and generating a plan you review and approve before any code changes are made.
Google describes Antigravity as “our agent-first development platform for developers to take an idea and turn it into a production-ready app.” Antigravity 2.0 is described as “a central home for agent interaction, allowing you to orchestrate multiple agents to execute tasks in parallel”, with “dynamic subagents for parallelized workflows, scheduled tasks for background automation and ecosystem integrations.”
The digital cities and games framing is Google’s own, on the AI products page: “Build a digital city in Google Antigravity using subagents working together” and “Build games that improve themselves in Google Antigravity.”
Google’s Stitch update post (19 May 2026) describes “transforming your design experience into a live, collaborative partnership with the Stitch Agent, allowing you to design, stream and steer iterations in real time”, accepts text prompts, voice, existing codebases and design files, and names three export routes directly: generate a shareable link via Google AI Studio, export screens into Google Antigravity to plug in backend logic, or publish to the web with Netlify.
Figma is the fourth export route named in the script. It is documented at Stitch’s launch rather than in the May 2026 update: designs “can be seamlessly pasted to Figma”.
Google’s help page states: “You can currently use Pomelli at no charge” and “Users can create several hundred image and video generations.”
The same page carries the warning quoted in the script: “AI can sometimes modify fine details of your products, like colors, logos, or textures. We recommend reviewing all AI-generated images before publishing to ensure they accurately represent your real-world products.”
Business DNA is Pomelli’s brand profile built from a business website, covering brand fonts, colours, values and images alongside the Pomelli Catalog and assets.
Google’s Flow update post (19 May 2026) states Flow “evolved into an AI creative studio, with new capabilities in video and image generation and editing, and launched in over 140 countries around the world”, and introduces Flow Agent as a partner that can “plan and reason through complex tasks with your inputs, under your control”, covering brainstorming, creation, editing, variations, batch editing and organising assets.
The same post states Flow Music incorporates Lyria 3 Pro, and describes granular editing: users can “highlight any part of a song and make changes” without affecting the rest of the track, “instantly rewrite or translate lyrics”, “restyle the beat drop”, a covers feature that transforms the style of full songs while keeping the original melody and structure, and directing shareable music videos conversationally.
Google’s support pages document creating songs from text, image and audio prompts, and editing, remixing, publishing, sharing by link and downloading them individually or as a zip.
Google describes Opal as a way to “build and share powerful AI mini apps that chain together prompts, models, and tools, all using simple natural language and visual editing”, with a visual node based editor, hosting, and sharing so others can run the app with their own Google account.
Google describes Mixboard as “an experimental, AI-powered concepting board designed to help you explore, expand and refine ideas”, with an open canvas combining text, images and generative AI, natural language editing, and one click regenerate and more like this options. It launched in public beta in the United States.
Google’s AI products page carries the line quoted in the script: “Create, edit and share videos, using new AI capabilities, at no cost in Google Vids.”
Google’s own Workspace post is titled “Create, edit and share videos at no cost in Google Vids”, and Google states anyone with a Google account can generate video clips at no cost using Veo 3.1.