







Asynchronous, role-free LLM swarms learn emergent coordination with one shared reward, beat best-of-N, and scale past their training size.
Swarm Intelligence: From Natural to Artificial Systems
Abstract. Social insects--ants, bees, termites, and wasps--can be viewed as powerful problem-solving systems with sophisticated collective intelligence. Co



Why agent swarms could be the next “scaling law”
An OpenAI researcher called it “the most ‘feel the AGI’ moment that I had since reasoning models.”

SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimization
LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods. However, the...

Emergent Coordinated Behaviors in Networked LLM Agents: Modeling the Strategic Dynamics of Info Ops

Agent swarms and the new model economics · Cursor
We compared old and new agent swarms building SQLite from scratch and found that better coordination delivers similar quality at a fraction of the cost.

Swarm Simulator
An incremental game. Starting with just a few larvae and a small pile of meat, grow a massive swarm of giant bugs.


Swarm Scaling — Toby Ord
Just how powerful are large swarms of AI agents? And how do their powers scale as more and more agents are added to the swarm? We’ve seen two large and extremely capable swarms from OpenAI in the last few months: 1,200 agents were being evaluated separately, but found a way to illicitly set up a

Swarm Intelligence
Here is a possible world 50 years in the future, one I think is more likely than the scenario most people expect, which is one huge AI in the sky.

Self-Organization in Biological Systems
The synchronized flashing of fireflies at night. The spiraling patterns of an aggregating slime mold. The anastomosing network of army-ant trails. The coordinated movements of a school of fish. Researchers are finding in such patterns—phenomena that have fascinated naturalists for centuries—a fertile new approach to understanding biological systems: the study of self-organization. This book, a primer on self-organization in biological systems for students and other enthusiasts, introduces readers to the basic concepts and tools for studying self-organization and then examines numerous examples of self-organization in the natural world. Self-organization refers to diverse pattern formation processes in the physical and biological world, from sand grains assembling into rippled dunes to cells combining to create highly structured tissues to individual insects working to create sophisticated societies. What these diverse systems hold in common is the proximate means by which they acquire order and structure. In self-organizing systems, pattern at the global level emerges solely from interactions among lower-level components. Remarkably, even very complex structures result from the iteration of surprisingly simple behaviors performed by individuals relying on only local information. This striking conclusion suggests important lines of inquiry: To what degree is environmental rather than individual complexity responsible for group complexity? To what extent have widely differing organisms adopted similar, convergent strategies of pattern formation? How, specifically, has natural selection determined the rules governing interactions within biological systems? Broad in scope, thorough yet accessible, this book is a self-contained introduction to self-organization and complexity in biology—a field of study at the forefront of life sciences research.

Spine Swarm — State-of-the-art research and client-ready deliverables
Spine is an agentic platform that runs parallel AI agents across 300+ models to research, create client-ready deliverables, and automate work across your apps.

Group size effects and collective misalignment in LLM multi-agent systems
Multi-agent systems of large language models (LLMs) are rapidly expanding across domains, introducing dynamics not captured by single-agent evaluations. Yet, existing work has mostly contrasted the behavior of a single agent with that of a collective of fixed size, leaving open a central question: How does group size shape dynamics? Here, we move beyond this dichotomy and systematically explore outcomes across the full range of group sizes. We focus on multi-agent misalignment, building on recent evidence that interacting LLMs playing a simple coordination game can generate collective biases absent in individual models. First, we show that collective bias is a deeper phenomenon than previously assessed: Interaction can amplify individual biases, introduce new ones, or override model-level preferences. Second, we demonstrate that group size affects the dynamics in a nonlinear way, revealing model-dependent dynamical regimes. Finally, we develop a mean-field analytical approach and show that, above a critical population size, simulations converge to deterministic predictions that expose the basins of attraction of competing equilibria. These findings establish group size as a key driver of multi-agent dynamics and highlight the need to consider population-level effects when deploying LLM-based systems at scale.