#model-weights

The latest AI and developer news about #model-weights, with the original source and a short summary.

Percepta introduces Spotlight architecture to separate intelligence from memory

Spotlight enables infinitely growing memory without increasing access costs by using an arbitrarily sparse indexing mechanism.

·2026.10.03 02:00

Proposal for 'Infinite-Parameter LLM' Architecture That Generates and Adapts Weights with Real-Time Data

An 'Infinite-Parameter LLM' architecture has been proposed, which converts real-time interaction data into weights via a hypernetwork and performs online updates using a Bayesian approach.

·1·2026.09.18 01:00

Laguna S 2.1 Weights Updated

Poolside updated the FP8 and NVFP4 checkpoints for Laguna S 2.1.

·2026.08.01 22:00

Cursor's New Model (1 min read)

Explains the concept and role of Model Weights, a core component of LLMs, through an analogy.

·2026.06.18 09:00

SenseNova-U1 Model and Training Code Released

SenseNova-U1 model weights and InternEvo-based training code have been open-sourced.

·1·2026.06.11 01:00