A coding model learns to make watercolor-style images
An open training recipe links generated drawing code to a visual reward loop.
What’s new, what matters, and what to build next.
An open training recipe links generated drawing code to a visual reward loop.
The project addressed the gap between keeping old sessions and retrieving useful context from them.
A public recipe measured whether light post-training could improve a compact model’s schema compliance.
A new JavaScript library and versioned kernel collection targeted the operations underneath browser inference.
The Monsoon evaluation sets broaden language coverage and make it possible to investigate differences hidden by an overall error rate.
Multi-vector fine-tuning offers a route to domain-specific retrieval, but good relevance data and evaluation matter as much as the model.
Typed nodes and visible intermediate outputs made multi-step AI apps easier to explore.
Speech-recognition evaluation needs to distinguish faithful listening from patterns learned around a familiar benchmark.
The engineering account connects exact matching, semantic retrieval, and Hub infrastructure.
Late-interaction models preserve finer matching information at the cost of larger indexes.
Explore releases and guides from 2024 onward. Dates refer to the original announcements; each article also shows when our coverage was published. How we cover the ecosystem ↗