Efficient Pretraining with Token Superposition

When large language models (LLMs) hallucinate, developers typically assume the model lacks the required facts. Engineering teams diagnose the error as missing knowledge. The standard response is to increase model size, expand training data, or build complex retrieval architectures.A new study by res

Sep 1, 2026

Earlier this year, we ran a pilot giving external researchers access to aggregate, real-world Claude usage data. Three research groups designed their own studies for Anthropic Insights, our privacy-preserving analysis tool. In this post, we share high-level results from those studies and what we learned running this pilot.

Aug 27, 2026

Nous Research launches Psyche, a distributed training network using the DisTrO optimizer and Solana blockchain to coordinate unbiased AI model training across the internet.

May 31, 2025

Nous Research introduces Lighthouse Attention, a selection-based hierarchical attention mechanism that runs ~17x faster than standard attention at 512K context and delivers 1.4-1.7x end-to-end pretraining speedup.

May 9, 2026