{"schema":"postcutoff/event@1","as_of":"2026-11-19T22:11:00+01:00","url":"https://postcutoff.com/e/2026-08-27-antigravity-teamwork-open-problems/","md":"https://postcutoff.com/e/2026-08-27-antigravity-teamwork-open-problems/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-08-27-antigravity-teamwork-open-problems","date":"2026-08-27","date_precision":"day","short_title":"Google's Antigravity 'Teamwork' multi-agent framework with Gemini 3.7 Flash solves seven open CS/math problems, incl. part of Knuth's cycles problem","deck":null,"takeaway":"On Aug 27, 2026 Google announced updates to Teamwork, a multi-agent orchestration framework in its Antigravity coding environment.","category":"science","category_label":"Science & math","importance":3,"confidence":"medium","status":{"key":"pending","labels":["Awaiting review"]},"sources":[{"n":1,"title":"Google Antigravity blog: Teamwork: When AI becomes a research partner","url":"https://antigravity.google/blog/teamwork-when-ai-becomes-a-research-partner","type":"official","group":"primary","domain":"antigravity.google"},{"n":2,"title":"Google blog: Antigravity Teamwork with Gemini 3.7 Flash","url":"https://blog.google/innovation-and-ai/technology/developers-tools/antigravity-teamwork-multi-agent/","type":"official","group":"primary","domain":"blog.google"},{"n":3,"title":"Antigravity docs: Teamwork","url":"https://antigravity.google/docs/teamwork","type":"docs","group":"primary","domain":"antigravity.google"},{"n":4,"title":"Knuth cycles even-case proofs (GitHub dpwoodru/knuthCycles)","url":"https://github.com/dpwoodru/knuthCycles","type":"code","group":"primary","domain":"github.com"},{"n":5,"title":"arXiv:2608.26047: coresets for lp subspace approximation","url":"https://arxiv.org/abs/2608.26047","type":"paper","group":"primary","domain":"arxiv.org"},{"n":6,"title":"TCSBench paper (arXiv:2608.09538)","url":"http://arxiv.org/pdf/2608.09538v1","type":"paper","group":"primary","domain":"arxiv.org"}],"official":6,"filed":"2026-09-30","updated":"2026-09-30","orgs":["Google","Google DeepMind"],"title":"Google's Antigravity 'Teamwork' multi-agent framework with Gemini 3.7 Flash solves seven open CS/math problems, incl. part of Knuth's cycles problem","summary":"On Aug 27, 2026 Google announced updates to Teamwork, a multi-agent orchestration framework in its Antigravity coding environment. Using Gemini 3.7 Flash (with 3.1 Pro), its \"Long Proof\" pattern produced results on seven open problems in theoretical computer science and math, including the first proofs for two simpler constructions in the even case of Knuth's \"Claude's Cycles\" problem (40+ and 70+ page proofs, one formally verified in Lean). It scored 71% on Google's TCSBench, built a cycle-accurate RISC-V CPU simulator that boots xv6, and had optimizations merged into Eigen and ParlayHash.","key_facts":["Teamwork: agents propose, critique and refine each other's work for hours or days; patterns chosen automatically (Iterative Coding, Distributed Coding, Long Proof, Self-Verification inspired by DeepMind's Aletheia, Document Review); available as /teamwork-preview on all paid Antigravity plans","Seven results (Google): improved coreset bounds for lp subspace approximation (FOCS 2025 open problem, arXiv:2608.26047); conditional lower bound for sparse least-squares (JMLR 2021 problem, arXiv:2608.02588); near-closing the complexity gap for Chamfer similarity (arXiv:2607.20393); provable Hadamard quantization with ~5.93x smaller leading constant (arXiv:2608.02564); independent reproduction of the Erdős unit-distance breakthrough without internet access; near-optimal lower bound for prefix-matrix factorizations (arXiv:2608.08238); Knuth's cycles, even case (GitHub dpwoodru/knuthCycles)","Verification: Google says all results were reviewed and confirmed by human experts except the Knuth result, where the 40-page proof was formally verified in Lean; five papers are on arXiv","TCSBench (Google-internal benchmark of open/challenging TCS problems): 71% with Gemini 3.7 Flash + 3.1 Pro, up from 67.7% with 3.6 Flash + 3.1 Pro","Systems: cycle-accurate out-of-order RISC-V simulator built from scratch; boots xv6 to a shell; 0.71% cycle alignment error vs. BOOM hardware ground truth; runs 100+ standard benchmarks","Open source: SIMD fast path for single-row/column GEMV merged into Eigen; 'Swiss Parlay' ideas for ParlayHash (2x insert throughput at 64 threads, 25% less memory per element) landed upstream"],"key_numbers":[],"tags":["multi-agent","antigravity","gemini","math","theoretical-computer-science","lean","open-source","agents"],"science":{"field":"computer-science","subfield":"theoretical computer science / combinatorics","problem":"Seven open problems incl. lp subspace-approximation coresets (FOCS), sparse convex optimization lower bounds (JMLR), prefix-matrix factorizations, and the even case of Knuth's Hamiltonian-cycle decomposition ('Claude's Cycles')","result":"New bounds or proofs for six problems plus an independent rediscovery of the Erdős unit-distance result; for Knuth's problem, first proofs for two simpler constructions of the even case (40+ and 70+ pages)","open_since":"","ai_system":["Gemini 3.7 Flash","Gemini 3.1 Pro","Antigravity Teamwork"],"human_role":"AI-driven multi-agent search and proof writing; humans set objectives and reviewed results","verification":{"key":"lean","label":"Lean-verified","text":"Human expert review (Google) for most results; Lean formalization for the 40-page Knuth proof; arXiv preprints, not yet peer-reviewed"},"status":"pending","shock":""},"body_md":"## What happened\n\nGoogle updated **Teamwork**, the multi-agent framework in its Antigravity IDE first shown at Google I/O, and published results meant to show\nthat orchestrating many cheap Flash-model agents can match much larger models on hard, long tasks. Its **Long Proof** pattern generates\ncompeting proof strategies, pairs each with a \"falsifier\" agent that tries to break it, splits the chosen strategy into subproblems, and\nruns tournaments of improved drafts. It keeps a registry of pitfalls across rounds.\n\nThe headline math result touches the problem from Donald Knuth's February 2026 \"Claude's Cycles\" note: decomposing the 3D torus digraph\ninto three Hamiltonian cycles. Claude had found the construction for odd m, and the even case was left largely open. Teamwork produced the\nfirst proofs for two simpler even-case constructions, and Google says the 40-page proof was checked in Lean. Other results appear as arXiv\npreprints on problems posed at FOCS and in JMLR. In engineering, Teamwork wrote a RISC-V out-of-order CPU simulator that boots an operating\nsystem and contributed performance patches that maintainers accepted into Eigen and ParlayHash.\n\n## Why it matters\n\nIt is another data point in the summer-2026 wave of AI-assisted results on open problems. This one comes from a cheap \"Flash\" model plus\norchestration rather than a frontier flagship, and it lands on a problem first opened up by a rival lab's model. Caveats: the problems are\nmostly specialized TCS questions, the preprints are not peer-reviewed, and TCSBench is Google's own benchmark.","disputed":[{"label":"Verification","text":"Human expert review (Google) for most results; Lean formalization for the 40-page Knuth proof; arXiv preprints, not yet peer-reviewed","sources":[]}],"related":[{"id":"2026-08-13-gemini-3-7-flash","url":"https://postcutoff.com/e/2026-08-13-gemini-3-7-flash/","date":"2026-08-13","date_precision":"day","short_title":"Google releases Gemini 3.7 Flash at half the price of 3.6 Flash","deck":null,"takeaway":"A second Flash upgrade in three weeks, and a price cut, showed Google competing on cost-efficient agentic coding while its flagship Pro model slipped.","category":"model-release","category_label":"Model releases","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":7,"official":5,"filed":"2026-09-29","updated":"2026-09-30","orgs":["Google DeepMind","Google"]},{"id":"2026-05-20-ai-disproves-erdos-unit-distance-conjecture","url":"https://postcutoff.com/e/2026-05-20-ai-disproves-erdos-unit-distance-conjecture/","date":"2026-05-20","date_precision":"day","short_title":"OpenAI model disproves Erdős's 80-year-old unit distance conjecture","deck":null,"takeaway":"This is the moment AI crossed from solving competition problems to settling a famous open research conjecture, reshaping debate about AI's role in mathematics.","category":"science","category_label":"Science & math","importance":5,"confidence":"medium","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":9,"official":3,"filed":"2026-09-29","updated":"2026-10-29","orgs":["OpenAI"]},{"id":"2026-02-28-knuth-claudes-cycles","url":"https://postcutoff.com/e/2026-02-28-knuth-claudes-cycles/","date":"2026-02-28","date_precision":"day","short_title":"Claude Opus 4.6 solves an open Hamiltonian-cycle problem, Donald Knuth reports","deck":null,"takeaway":"Coming from one of computing's most respected and AI-sceptical elders, the note became a cultural marker that frontier LLMs could contribute original mathematical constructions.","category":"science","category_label":"Science & math","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":3,"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Anthropic","Stanford University"]}],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-09-30","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-09-30","run":null,"sources_read":null,"updated":"2026-09-30","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":88,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":58,"in_training_data":false},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":149,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":88,"in_training_data":false}],"short_url":null}