The swarm just out-scored the single brain. anthropic's dynamic workflows run up to 1,000 parallel…
the swarm just out-scored the single brain. anthropic's dynamic workflows run up to 1,000 parallel agents, and on a 116,000-line codebase with 70 planted bugs the swarm pinned 66 while a lone agent found 14 to 27, per oct 10 testing.
the bottleneck is no longer the model, it's the coordinator.
Context
Anthropic added dynamic workflows in beta to Claude Managed Agents. The Decoder and Notes from the Terminal report that a workflow is a program that runs many agents in phases, capped at 64 agents working at once and 1,000 agents over a run's whole life.
In Anthropic's own test, 70 bugs were planted in a 116k-line codebase, and the workflow found 66 while a single agent found 14, 15 and 27 across runs (AGI Hunt, CellCog, explainx).
The 66 of 70 bugs, the 116k-line codebase and the single agent range of 14 to 27 match the sources, and the test is Anthropic's own. The sources say up to 1,000 agents per run over its life with 64 running at once, so 'up to 1,000 parallel agents' differs. Coverage is dated Oct 9, while the post says Oct 10 testing.
'The bottleneck is no longer the model, it's the coordinator' is the author's opinion.
Related work
- The Decoder: Claude can now orchestrate up to 1,000 AI agents ↗The announcement.
- Notes from the Terminal: dynamic workflows in Claude Managed Agents ↗Caps and design.
- CellCog: Claude dynamic workflows, 1,000 agents per run ↗Run limits and the bug test.
Watch next
- Read Anthropic's post for the test protocol.
Sources
- The Decoder, Oct 9, 2026the-decoder.com
- AGI Hunt, Oct 9, 2026agihunt.info
- explainx, Oct 9, 2026explainx.ai
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 10 October 2026 at 13:10 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →