Google Antigravity

Headlines · 8

1. 1d
   Google Antigravity agents speed up antimicrobial peptide discovery; lead enters wet-lab tests

   In an October 15 technical blog, Google DeepMind described how researchers used Google Antigravity, its agentic development platform, to accelerate antimicrobial peptide (AMP) discovery: the scientist sets the biological goal and steers strategy while the agent handles cross-database orchestration, structural analysis and literature synthesis, compressing days of fragmented database triage into a single session. The platform uses ProtNLM, a deep learning model that predicts biological function directly from amino acid sequences, to enrich millions of uncharacterized UniProt entries with high-confidence functional labels, and adds an embedded Biomolecular Structure Viewer that gives researcher and agent a shared 3D canvas for annotation and structure-based questions. The post says the team uncovered a novel AMP candidate in Streptococcus pseudopneumoniae with predicted activity against pathogens involved in respiratory tract infections, and that the lead plus several others from the same pipeline are now in wet-lab validation.

   Google DeepMind Blog · 🔥 10
2. 2d
   Gemini 4 Argon opens globally, targeting robotics and enzyme design

   Google DeepMind opened Gemini 4 Argon globally on October 13, after debuting it September 30 for selected cybersecurity defenders; access now runs through Google AI Pro and Ultra subscriptions, the Gemini API, Google AI Studio and Antigravity, while free users are excluded. Introductory API pricing is $2 per million input tokens and $10 per million output tokens through November 30, doubling to $4 and $20 on December 1. Google pitches Argon as a general physical-reasoning model for the whole robotics pipeline, building MuJoCo simulation twins from a single photo, generating 4,500-plus training episodes, lifting real-robot task success from 56% to 75%, and in a Boston Dynamics Spot demo using extra agents to verify foot and gripper placement. It also says researchers used the model to design five catalytically active serine esterases in hours. These are Google's own claims: benchmark accounts diverge, with some reporting leader-level Vals AI scores and others a 53-point Artificial Analysis Intelligence Index, weaker terminal-control and scientific-reasoning results, and robotics demos that still needed detailed human prompting.

   钛媒体 · 🔥 40
3. 3d
   Google Antigravity is now built into the Gemini app for macOS

   Alongside the Gemini 4 Argon rollout, Google has integrated Antigravity directly into the Gemini app for macOS. A Liquid Glass button in the top-right corner launches the same full experience found in the standalone app, letting users build apps and web experiences from one place. The report says the new model requires AI Pro or above in the Gemini app, while Antigravity access requires AI Ultra. The macOS app also gains a Chat/Task switcher (Task is the new name for Gemini Spark), a Scheduled view, and @ mentions for Skills instead of the slash command, with Connectors replacing Connected Apps.

   9to5Google · 🔥 0
4. 4d
   Gemini 4 Argon cut ADK framework overhead by 40% in an 11-hour unattended run

   On the launch of Gemini 4 Argon, Google engineers let the model run unattended against google/adk-python, a roughly 205,000-line, nearly 800-file repository backed by almost 17,000 unit tests. A first two-hour attempt found real bottlenecks but ignored maintainer conventions; a second round using in-repo .agents/skills rulefiles and Antigravity's /goal mode ran 11 hours overnight and proposed six optimizations across 32 files. After code review only two shipped: filtering in-memory session history before copying, and lazy-loading heavy packages such as streaming audio/video, OpenTelemetry and FastAPI/Starlette. Google's own independent holdout benchmark reports roughly 40% less framework overhead in long sessions (a 158-turn run fell from 65.6s to 56.1s) and 21% faster cold imports of google.adk, while time from a fresh process to first model request stayed flat; the other four changes were rejected as fragile or synthetic-only.

   Google Developers Blog · 🔥 0
5. 5d
   Google releases Gemini 4 Argon for developers and Gemini app users

   Google released Gemini 4 Argon on Oct 13, making the frontier model available to developers and enterprises via the paid Gemini API in AI Studio and the Gemini platform, to Google AI Ultra subscribers in Antigravity, and to AI Ultra and Pro subscribers in the Gemini app and AI Mode in Search. Google says Argon targets long-horizon work such as code migration, enterprise knowledge work and cybersecurity defense, ranking #1 on the Vals Index, claiming 77.9% on DeepSWE and citing more than 35% higher agentic code submissions in Antigravity, though its own comparison table shows it trailing Claude Opus 5.5 and GPT-6 Astra on some coding and terminal tests. Introductory pricing is $2 per million input tokens and $10 per million output tokens, with cached input at 95% off, and Manus says Argon is now live on its platform.

   TestingCatalog · 🔥 0
6. 7d
   K-Dense's Scientific Agent Skills library hits 177 skills, adds AI co-scientist app

   K-Dense's Scientific Agent Skills collection — formerly Claude Scientific Skills — has been renamed and rebuilt around the open Agent Skills standard, so it now targets any compatible host including Cursor, Claude Code, Codex and Google Antigravity, not just Claude. Version 2.72.0 ships 177 skills spanning bioinformatics, cheminformatics and drug discovery, medicine, materials, physics, Earth science and scientific writing, with a database skill covering 80 sources and 70-plus version-aware Python package workflows. The project also launched K-Dense BYOK, a free open-source desktop "AI co-scientist" where users bring their own API keys, pick from 40+ models, keep data local and optionally scale to Modal, and it claims 250,000+ scientists already use the library.

   GitHub Trending · Python · 🔥 0
7. 7d
   Free Gemini app users limited to 'Auto' model picker as thinking levels roll out

   Google is rolling out a change to the Gemini app for users on the free plan: those without a plan can no longer switch models, and the only option in the model picker is "Auto" with "Gemini 3 models." Google says most prompts will go to the fastest model, Flash-Lite, while prompts needing deeper reasoning may go to Flash or Pro; if smart model selection is off, all chats for users without a plan use Flash-Lite. New thinking levels (Low, Medium, High) replace the previous "Extended" option and are available to all subscribers, matching Google Antigravity and AI Studio; per 9to5Google, Google's support documentation referred to the free-tier model as (3.5) Flash-Lite, and Google AI Plus users will soon be limited to Flash-Lite and (3.8) Flash.

   9to5Google · 🔥 0
8. Oct 7
   Google launches Developer Knowledge API ecosystem for AI agents and dev tools

   Google introduced the Developer Knowledge API ecosystem, a structured programmatic interface to official documentation for Google Cloud, Firebase, Android and more, aimed at AI agents, IDE extensions and automated workflows rather than web scraping. The lineup includes gcloud CLI commands (answer-query, documents search-chunks, documents describe), an official agent skill that connects via MCP to Google Antigravity, Claude Code, Cursor, GitHub Copilot and custom frameworks, client libraries for C#, Go, Java, Node.js/TypeScript, PHP, Python and Ruby, and an APIs Explorer for interactive testing. Google says the API is indexed frequently to keep retrieval fresh and reduce hallucinations caused by stale LLM training cutoffs.

   Google Developers Blog · 🔥 0

Experience and discussion from the community

Share my Google Antigravity experienceAsk about Google Antigravity