Google @Antigravity
A Google DeepMind program for 10,000 scientists that is expanding its access to Gemini 4 Carbon. It appears to be an initiative around scientific AI access and experimentation.
Key Highlights
- Google @Antigravity is positioned as a Google DeepMind environment for scientific AI access, experimentation, and workflow support.
- The platform evolved from launching Science Skills in May 2026 to expanding free Gemini 4 Carbon access for 10,000 scientists in November 2026.
- Its compatibility with reusable scientific-agent workflows points to an ecosystem strategy rather than a single-purpose assistant.
- For AI PMs, Antigravity is a strong example of vertical AI product packaging that combines models, skills, and targeted user acquisition.
- Newsletter mentions connect Antigravity to adjacent tools like Claude Code, Cursor, Codex, and Gemini Live APIs.
At a glance
- Category
- Agent platform
Google @Antigravity
Overview
Google @Antigravity is a Google DeepMind program and tool environment focused on giving scientists access to AI-powered research workflows. Based on the newsletter mentions, it combines model access, domain-specific skills, and integrations that support experimentation across scientific and adjacent use cases. Over time, it has been described as the home for "Science Skills," a host for reusable scientific-agent workflows, and a program through which 10,000 scientists received free access to high-usage tiers tied to Gemini model access.For AI Product Managers, Google @Antigravity matters because it shows how a frontier AI lab is packaging models, skills, and distribution into a targeted product for expert users rather than a general-purpose chatbot audience. It is a useful case study in vertical AI product design: pairing specialized data sources, reusable agent skills, and time-limited access programs to drive adoption in research settings. Its mentions alongside tools like Claude Code, Codex, Cursor, Gemini Live APIs, and Gemini 4 Argon/Carbon also suggest an ecosystem approach rather than a standalone model surface.
Key Developments
- 2026-05-22: Google DeepMind launched Science Skills for Google @Antigravity, an AI toolkit integrating insights from 30+ life-science sources such as UniProt and the AlphaFold Database to speed up day-to-day research workflows.
- 2026-07-14: Google DeepMind highlighted the "Predicting the Past" skill in Google Antigravity, using it to investigate a Roman ring thief, map an ancient European cult, and reconstruct visitor networks to a Greek oracle.
- 2026-10-06: Peter Yang demonstrated Tabi, a phone-friendly Japanese AI language coach built with Gemini Live APIs, Nano Banana, Google Antigravity, Claude Code, and Vercel, showing Antigravity in a lightweight application-building workflow.
- 2026-10-15: Google DeepMind announced that 10,000 scientists would receive free access for six months to the highest usage tier in Google Antigravity, positioned to support research with Gemini 4 Argon.
- 2026-10-21: K-Dense-AI/scientific-agent-skills noted compatibility with Google Antigravity alongside Cursor, Claude Code, and Codex, indicating Antigravity can act as a host environment for reusable scientific workflows.
- 2026-11-17: Google DeepMind expanded the Google Antigravity program for 10,000 scientists to include free access to Gemini 4 Carbon.
Relevance to AI PMs
1. A template for vertical AI packaging: Antigravity shows how to bundle foundation models with domain-specific skills and curated data access. AI PMs building for healthcare, finance, legal, or scientific users can study this pattern when designing products that need more than a raw chat interface.2. A distribution and adoption playbook: The six-month free access program for 10,000 scientists is a concrete example of seeding usage in a high-value expert segment. AI PMs can apply similar cohort-based rollout strategies to test retention, workflow fit, and model-tier economics before broader commercialization.
3. An interoperability lesson for agent products: Its appearance alongside Cursor, Claude Code, Codex, and scientific-agent-skills suggests users value portable workflows across hosts. AI PMs should think beyond model quality alone and prioritize compatibility, reusable skills, and ecosystem fit.
Related
- Google DeepMind: The organization behind Google @Antigravity and the source of its program announcements, scientific skills, and scientist access initiatives.
- Science Skills: A toolkit launched for Google @Antigravity that connects to 30+ life-science sources to support research workflows.
- Predicting the Past: A named skill used inside Google Antigravity for historical and archaeology-related investigations.
- Tabi: An example app built using Google Antigravity together with Gemini Live APIs, Claude Code, and Vercel, illustrating broader experimentation beyond pure scientific research.
- Gemini Live APIs: Part of the stack used in a project that also used Google Antigravity, suggesting complementary product surfaces.
- Claude Code, Cursor, Codex: Developer and agent-oriented tools mentioned as compatible hosts or adjacent workflow environments, useful for understanding Antigravity's ecosystem position.
- Gemini 4 Argon / Gemini 4 Carbon: Model access appears to be a major value layer in the Antigravity program, first with Argon-focused access and later expanded Carbon access.
- K-Dense-AI/scientific-agent-skills: A reusable library of scientific workflows that explicitly works with Google Antigravity, reinforcing its role as an execution environment for specialized research tasks.
Frequently asked questions
What is Google @Antigravity?
A Google DeepMind program for 10,000 scientists that is expanding its access to Gemini 4 Carbon. It appears to be an initiative around scientific AI access and experimentation.
What should AI product managers know about Google @Antigravity?
Google @Antigravity is positioned as a Google DeepMind environment for scientific AI access, experimentation, and workflow support. The platform evolved from launching Science Skills in May 2026 to expanding free Gemini 4 Carbon access for 10,000 scientists in November 2026. Its compatibility with reusable scientific-agent workflows points to an ecosystem strategy rather than a single-purpose assistant.
Newsletter Mentions (6)
“#6 𝕏 Google DeepMind announced that the Google Antigravity program for 10,000 scientists is expanding to include free access to Gemini 4 Carbon.”
#6 𝕏 Google DeepMind announced that the Google Antigravity program for 10,000 scientists is expanding to include free access to Gemini 4 Carbon.
“Each skill covers a package, data source, or workflow, and the library works with compatible hosts including Cursor, Claude Code, Codex, and Google Antigravity.”
#13 🐙 K-Dense-AI/scientific-agent-skills K-Dense-AI/scientific-agent-skills - K-Dense-AI/scientific-agent-skills is an MIT-licensed library of 177 reusable research skills and access guidance for 100+ scientific databases, spanning biology, medicine, engineering, Earth science, data analysis, and more. Each skill covers a package, data source, or workflow, and the library works with compatible hosts including Cursor, Claude Code, Codex, and Google Antigravity. Worth trying if you need portable scientific workflows for an AI agent.
“Google DeepMind announced that 10,000 scientists will receive free access for six months to the highest usage tier in Google Antigravity, aiming to advance research with Gemini 4 Argon.”
#3 𝕏 Google DeepMind announced that 10,000 scientists will receive free access for six months to the highest usage tier in Google Antigravity, aiming to advance research with Gemini 4 Argon. #4 🐙 morluto/rea morluto/rea - **morluto/rea** is an MIT-licensed MCP server and command-line toolkit for locally inspecting native binaries, JavaScript/Electron apps, .NET assemblies, and websites with supporting evidence. morluto released rea-agents 6.5.0, dated 2026-10-13 and published 2026-10-14T12:32:45Z, adding scoped JavaScript analysis, cross-bundle recovery, retained evidence views, and offline artifact inspection.
“Peter Yang demonstrated Tabi, a phone-friendly Japanese AI language coach built in six steps using Gemini Live APIs, Nano Banana, Google Antigravity, Claude Code, and Vercel.”
#6 ▶️ Build an AI Language Tutor You Can Talk To (6 Steps) Peter Yang Peter Yang demonstrated Tabi, a phone-friendly Japanese AI language coach built in six steps using Gemini Live APIs, Nano Banana, Google Antigravity, Claude Code, and Vercel. Its 10 lessons teach 10 travel phrases each before role-play, and the app took about three hours to build and about five minutes to deploy on Vercel.
“Google DeepMind used the “Predicting the Past” skill in Google Antigravity to track down a Roman ring thief, map an ancient European cult, and reconstruct networks of visitors to a Greek oracle.”
#15 𝕏 Google DeepMind used the “Predicting the Past” skill in Google Antigravity to track down a Roman ring thief, map an ancient European cult, and reconstruct networks of visitors to a Greek oracle.
“Google DeepMind launched Science Skills for Google @Antigravity, a new AI toolkit integrating insights from 30+ life-science sources—like UniProt and the AlphaFold Database—to accelerate day-to-day research workflows.”
#8 𝕏 Google DeepMind launched Science Skills for Google @Antigravity, a new AI toolkit integrating insights from 30+ life-science sources—like UniProt and the AlphaFold Database—to accelerate day-to-day research workflows.
Related
Anthropic’s coding agent/product mentioned as a supported client for the ffmpeg-skill repository. It is cited as part of an offline video-editing workflow for agents.
An AI coding environment mentioned as a client supported by ffmpeg-skill. It appears here as part of an agentic tooling stack for local video editing.
An AI coding tool mentioned as a supported client for ffmpeg-skill. It is listed with Claude Code and Cursor in the newsletter.
Google's AI research organization. It is highlighted for multiple developer launches and for expanding the Google Antigravity program with Gemini access.
A Gemini model variant mentioned as part of Google DeepMind's developer launches and later as being added to the Google Antigravity program. It appears to be a notable model offering for scientists and developers.
Stay updated on Google @Antigravity
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free