Google's Cogentic lets Gemini find new proofs for 5 unsolved math problems via multi-agent orchestration
rohanpaul_ai · x · 2026-10-02
A new Google paper presents Cogentic, a system that organizes Gemini like a research team: multiple agents explore different ideas in parallel, checkers treat every step as wrong until proven, and validated results are stored in shared notes for later rounds.
Most problems took only about 100 model calls, and human experts confirmed every proof. The takeaway: for long, hard agent tasks, pair your agents with a strict checker and a running record of proven work rather than relying on a better prompt alone.
Related event: Google's Cogentic Multi-Agent System Tackles Open Math Problems(4 posts)→
More from coding & agent
- Claude Code overtakes Cursor and Copilot in job-post mentions ahead of Anthropic IPO — antgoldbloom · 2026-11-19
- Dev submits PR adding periodic checkpoints to GKE sandboxes so agents don't lose progress — yoimnotkesku · 2026-11-19
- AWS adds private CA support to Bedrock AgentCore Identity for connecting agents to self-hosted IdPs — AWS ML Blog · 2026-11-19
- Orchestrator's Cloud Agents graduate from beta to general availability — julianweisser · 2026-11-19
- Zero users, 33 pages shipped: dev crowdsources audit for AI-agent trust schema — TheHamzaShahbaz · 2026-11-19
- She vibecoded her own photobooth with Claude Code — and shared the prompt — toolstelegraph · 2026-11-19