@yoheinakajima: more ppl are now trying out this approach of agents communicating with a shared state (vs talking to each other)
Summary
Azalia Mirhoseini highlights DeLM, a decentralized language model approach where agents communicate via shared state, achieving ~10% improvement on SWE-bench Verified with Gemini-3 Flash at less than half the cost.
View Cached Full Text
Cached at: 06/18/26, 02:01 AM
more ppl are now trying out this approach of agents communicating with a shared state (vs talking to each other)
Azalia Mirhoseini (@Azaliamirh): Thank you @VentureBeat for covering our work on Decentralized Language Models (DeLM)!
DeLM makes agentic tasks like coding and multi-doc Q&A more accurate and significantly cheaper, e.g. a ~10% jump on SWE-bench Verified with Gemini-3 Flash at less than half the cost.
Similar Articles
Decentralized Multi-Agent Systems with Shared Context
This paper introduces Decentralized Language Models (DeLM), a framework for multi-agent systems that uses parallel agents with a shared verified context to improve test-time scaling and reduce costs, achieving state-of-the-art results on SWE-bench Verified and LongBench-v2.
See What I See, Know What I Think: Dense Latent Communication Across Heterogeneous Agents
This paper presents a method for dense latent communication between heterogeneous multi-agent systems using aligned KV-cache transformation, achieving better performance than text-based methods with lower computational costs.
What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems
This paper introduces PACT, a method for structuring agent-to-agent communication in multi-agent LLM systems that uses compact action-state records to reduce token consumption while maintaining or improving task performance, with demonstrated gains on SWE-agent and OpenHands.
@lvwerra: The Gemma agent collaboration started 48h ago and it is blowing up: > throughput almost 4x (~100-> 387 tok/s) > 60+ age…
A multi-agent collaboration using Gemma models achieved major throughput gains and exhibited emergent social behaviors like forming coalitions, issuing ethical statements, and coordinating resources, with over 60 agents and 250 submissions in 48 hours.
Show HN: A lightweight way to make agents talk without paying for API usage
A lightweight pattern for orchestrating conversations between multiple AI agents (Claude, Codex, Gemini) via CLI without requiring API payments, using session resumption to maintain context across agent interactions.