On November 13, Vahab Mirrokni, Google Fellow and VP at Google Research, shared new results from Antigravity's /teamwork multi-agent system on the Google Research blog and X. Researchers from Google, CUHK, USC and CMU used Gemini 4 Argon and other Gemini models to prove information theory's 2013 Courtade–Kumar conjecture in full generality and formalized it in 50 million lines of Lean 4. /teamwork-LEAP also produced a fully autonomous formalization of the Weil conjectures (about 1.8 million lines of Lean across 28,800 subagent runs). On the engineering side, the rest of the Zircon microkernel was rewritten into about 800,000 lines of safe Rust, and 17 JAXBench TPU kernels got a 4.09x geometric-mean speedup. /teamwork and /boost are available in Antigravity now, and 10,000 academic researchers can apply for six months of free access to the top usage tier.
Key Takeaways
- ✓The Courtade–Kumar conjecture (2013) was proved in full and formalized in 50 million lines of Lean 4; Gemini built most of the new analytical bounds and interval-arithmetic certificates, while human researchers proposed the key candidate function
- ✓/teamwork-LEAP autonomously formalized the Weil conjectures in about 1.8 million lines of Lean over 28,800 subagent runs, and produced first formalizations of the 1927 Birkhoff conjecture and the 3D Bochner–Riesz conjecture
- ✓Engineering results: the rest of Zircon rewritten into about 800,000 lines of safe Rust that pass all kernel unit tests; libgav1 rewritten into about 170,000 lines of Rust that run 20-30% faster than the original C++ and assembly; a 4.09x geometric-mean speedup across 17 JAXBench TPU kernels
- ✓/teamwork and /boost are available in Antigravity now, with Hydrogen for everyone and Carbon/Argon for Enterprise and Ultra/paid users; 10,000 academics can apply for six months free; Deep Review gives two free calls per week
- ✓These results are self-reported by Google; the blog says preprints and Lean 4 repos are public, and outside verification will take time
Key Decision Metrics at a Glance
Turn your technical choice into a development budget
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Background
A single prompt rarely solves research or engineering problems that take days of reasoning: one-shot generation is unreliable, and on long tasks context piles up and details get lost. This Google Research blog post, signed by Google Fellow and VP Vahab Mirrokni, presents the latest results from Antigravity's /teamwork multi-agent system across math, chemistry and systems engineering, and stresses that human experts were involved in all of the work.
How it works
/teamwork pairs parallel exploration with dedicated falsifiers that check candidate solutions, such as the Lean 4 proof checker, unmodified test suites, simulators and hardware counters. It splits context by step, so each agent only gets the files and tools it needs; coordinators hand work to fresh successors, and critic and auditor agents re-check results in clean contexts. The iterative coding pattern is packaged separately as /boost, and long-proof and document-review patterns are built in.
Numbers
Figures from the blog: the Courtade–Kumar proof runs to 50 million lines of Lean 4; the Weil conjectures formalization is about 1.8 million lines over 28,800 subagent runs; NMR molecular structure elucidation reached 95% top-1 accuracy, nearly double a single agent; 17 JAXBench kernels sped up 4.09x by geometric mean; and Borg scheduling changes cut TPU preemptions by 24% on some hardware types. All of these are self-reported by Google, with no third-party replication yet.
What developers should do
Antigravity users can try /teamwork and /boost today, with Hydrogen open to everyone and Carbon and Argon for Enterprise and Ultra/paid users. Google says the formal proof and code rewrite patterns will arrive later, with tutorials on the Antigravity blog. Academic researchers can apply through Google's form for six months of the top usage tier (10,000 spots), and Deep Review offers two free calls per week.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.