--- format: "aidr-story-markdown/v1" id: "630e9900ecd319dfb8dd54278f38a6d4dec638eb46da410b484d9f09f6995391" canonical_url: "https://aidr.today/630e9900?lang=en" title: "Google Antigravity Agent Hits 71% on TCSBench in Shift to Research Partner" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-08-28T05:47:07.000Z" category: "Research" topics: ["google","gemini","agent","coding","benchmark"] source_urls: ["https://huggingnews.com/ai/google-antigravity-agent-hits-71percent-on-tcsbench-in-shift-to-research-08feaecb","https://x.com/mirrokni/status/2093107188814290947"] summary: "Google researchers upgraded the Gemini-powered Antigravity agent /teamwork to function as a research partner capable of high-performance coding and hardware simulation. The agent hit 71% on the TCSBench benchmark and solved multiple open problems in sparse convex optimization and subspace approximation. It also designed a cycle-level RISC-V CPU simulator that boots an OS to shell with less than 1% average error. The update transitions the agent from \"Thinking Graphs\" to \"Graph Engineering,\" treating long-horizon research as a dynamic execution graph rather than rigid prompt engineering. /teamwork delivered production code optimizations to major C++ libraries, including a 32x speedup in Eigen. Google is releasing new harness patterns for iterative coding and document review through the AGY platform." --- # Google Antigravity Agent Hits 71% on TCSBench in Shift to Research Partner > [Open the canonical story]() **Published:** 2026-08-28T05:47:07.000Z **Category:** Research **Topics:** google, gemini, agent, coding, benchmark ## Summary Google researchers upgraded the Gemini\-powered Antigravity agent /teamwork to function as a research partner capable of high\-performance coding and hardware simulation\. The agent hit 71% on the TCSBench benchmark and solved multiple open problems in sparse convex optimization and subspace approximation\. It also designed a cycle\-level RISC\-V CPU simulator that boots an OS to shell with less than 1% average error\. The update transitions the agent from "Thinking Graphs" to "Graph Engineering," treating long\-horizon research as a dynamic execution graph rather than rigid prompt engineering\. /teamwork delivered production code optimizations to major C\+\+ libraries, including a 32x speedup in Eigen\. Google is releasing new harness patterns for iterative coding and document review through the AGY platform\. ## Sources - [Story source]() - [Story source]()