Research formalizes 'diversity collapse': multi-agent LLM systems homogenize outputs due to structural coupling — brainstorming setups must be explicitly engineered for heterogeneity.
Meta's first model from its Superintelligence Labs scores 50.2 on HLE With Tools using parallel multi-agent inference at test time.
Anthropic's Project Deal experiment ran live agent-to-agent negotiations among 69 staff: 186 deals, $4,000+ volume, Opus outperformed Haiku — but participants couldn't tell the difference.
Anthropic ran a live two-sided agent marketplace with 69 employees: 186 deals, $4,000+ volume — and model quality (Opus vs Haiku) was invisible to human participants throughout.
Sakana AI's Fugu beta hits SOTA on SWE-Pro, GPQA-D, and ALE-Bench with dynamic frontier model orchestration via an OpenAI-compatible API.
Curated AI insights — sent when there's something worth your inbox.