SponsoredChatgpt/Claude/Grok 会员24h在线自助充值 安全快速 稳定保障
奥创(Ultra)商城会员充值中心
SponsoredChatgpt/Claude/Grok 会员24h在线自助充值 安全快速 稳定保障
奥创(Ultra)商城会员充值中心
Sponsored本页公开展示实时样本、模型覆盖与响应时间。先用 5 元按量跑自己的 Claude Code/Codex 任务,成功后再决定是否长期使用。
ACS Gateway
Sponsored【腾讯云】WorkBuddy 全场景 AI 办公工作台,团队旗舰版低至198元/月
腾讯云
Sponsored人人可用的AI应用创作平台 自然语言交互,全栈应用搭建,伟大创意快速实现;个人版新用户注册送5000积分
阿里云Google DeepMind says researchers are using Gemini 4 Carbon in Google Antigravity for multi-step work across physics, mathematics and the life sciences.
Google DeepMind says researchers are putting Gemini 4 Carbon to work in Google Antigravity on complex, multi-step problems in physics, mathematics and the life sciences. The model is designed for tasks that call for extended reasoning and verification.
Examples shared by Google include analyzing lab notebooks and raw experimental data to flag anomalies that may affect results. These cases illustrate how the model can support research analysis; they do not establish that it can replace expert review or experimental validation.
Source review does not guarantee permanent accuracy. Consider the original source, observation time, and later corrections.
## Put Gemini 4 Carbon to work with `/swarm` Working with a single agent works well for scoped tasks, but breaks down when a change spans dozens of files and test suites. Dynamic subagents evolved that into a primary agent spinning off isolated helper agents: powerful for delegating one-off tasks, but still bottlenecked by a single coordinator context reviewing, synthesizing, and routing every step. With `/swarm`, working with Antigravity turns into orchestrating a coordinated system of specialized agents that plan, build, and verify in parallel. [...] ·Nov 10, 2026·3 min read Share on XShare on BlueskyShare on LinkedIn Gemini 4 Carbon and /swarm Today, we’re rolling out Gemini 4 Carbon inside Google Antigravity alongside `/swarm` (in preview), pairing our strongest model for long-horizon software engineering, knowledge work, scientific discovery, and cybersecurity with native multi-agent team orchestration. ## Put Gemini 4 Carbon to work with `/swarm` [...] Right-sized multi-mode
## Stronger scientific and mathematical reasoning Gemini 4 Carbon shows strongly improved performance in reasoning across physics, math, and life sciences, synthesizing experimental data, testing hypotheses, and reasoning through multi-step domain problems. For drug toxicity research company Axiom Bio, Carbon is now the leading model in performance for translational science, finding novel mechanisms of toxicity they hadn’t found previously. Axiom is now moving their production scientific analyses and agents to both Carbon and Argon to unlock the creation of safer drugs. Researchers across Google have been using Carbon in Google Antigravity to accelerate their work: [...] Enzyme assay development: Provided with only lab notebooks and raw data, Carbon went beyond standard analysis to surface hard-to-interpret data anomalies, which it used to diagnose subtle wet lab experimental artifacts such as incomplete automated mixing in deep wells, leading to substantially improved reproducibilit
high-precision knowledge work, artifact generation, and data analysis engineering tasks that require multi-agent collaboration strong visual design, spatial reasoning, and attention to environmental detail strong performance in reasoning across physics, math, and life sciences cybersecurity defense, vulnerability discovery, and patching (with specialized capabilities available to verified defenders via the Fairwind Program) ### Known Limitations [...] Methodology: deepmind.google/models/evals-methodology/gemini-4-carbon ## Intended Usage and Limitations ### Benefit and Intended Usage Gemini 4 Carbon is built for tasks that test the limits of a model’s intelligence, such as complex, multi-step coding and debugging and open-ended scientific research and discovery — using more thinking tokens than Gemini 4 Argon to explore hypotheses thoroughly, reason deeply, and run extended verification loops, with configurable thinking levels to balance quality, latency, and cost. It is parti
Over the past several weeks, teams across Google and academic partner institutions have tackled open problems in math, science, and complex systems engineering that had remained open for years. Using Gemini 4 Argon, Gemini 4 Carbon, Gemini 4 Hydrogen and multi-agent workflows such as Antigravity’s /teamwork (building on various Gemini models), these collaborations have produced new innovations across information theory, formal verifications, molecular structure discoveries, and kernel optimization, along with several production infrastructure gains at Google. We are making these agentic tools available in Antigravity’s /teamwork (alongside /boost and other agents) so that researchers and developers can apply them to their own work. [...] ## Mathematics and theoretical computer science Across mathematics and theoretical computer science, researchers paired with Gemini 4 and /teamwork harnesses to collaborate on large-scale proofs and generate complete algorithms and proofs directly fro
[39:54] and will never be invoked because nobody can actually install them on their device or even load them in the web browser. [40:00] So take a look today in your code base, see what you can remove, and share how much code you were able to remove with Antigravity and Gemini 4 Argon. Thank you, Rody, for the demo. [40:09] That was so cool to see what you can do with Antigravity to improve your day-to-day workflow as a software engineer. [40:14] Yeah. Watching Gemini 4 work through a real project inside Antigravity makes everything that Tulsee said earlier about persistence really click. [40:24] So you can actually see the shift from an agent acting as an assistant that autocompletes a few lines to an agent that you can actually hand a multi-step engineering problem to. [...] In our main segment, Smitha sits down with Tulsee to take you inside the lab to share how Google dogfoods early checkpoints through internal experiments. We will also discover why Gemini 4 Argon (built for deep r
[0:34] Google built it for two things in particular. Cyber security and what they're calling long horizon work, which basically means you can give Carbon big [0:42] complicated tasks that require a lot of steps and a lot of time to finish and it will do it. If you look at the benchmarks for Carbon real quick, what's most surprising is how much better it is [0:50] than Argon, the model they released only about a month ago. In fact, it beats Argon across every single benchmark. So, I was very excited to get my hands on [0:59] this and start testing it and comparing it to Argon. And here are my results. [1:03] Now, I mostly tested Carbon inside of Google Anti-Gravity, which is a coding app kind of similar to Claude Code or Codeex if you're familiar [...] [15:21] of engineering in roles like product, marketing, finance, legal, and research started using anti-gravity every day for [15:28] regular work. So today, Google launched a dedicated knowledge work mode that strips out the remaining c