Guide
A practical guide to SLMs: main architectures, when they outperform frontier models on production tasks, and how to fine-tune one.
Read more
Engineering
Our serving path falls off a cliff at adapter 33. Here's why, and what we did about it.
Read more
Research
The research behind Pioneer, where fine-tuning small language models becomes a closed loop.
Read more
Product
We’re launching Pioneer, the world’s first agent for fine-tuning and inferencing open source SLMs and LLMs.
Read more
Research
The future belongs to models with architectures crafted, optimized, and deployed for focused tasks.
Read more
Research
The one-size-fits-all LLM era is over. Scaling AI effectively now demands matching models to the right domain or task.
Read more
Correcting Stochastic Update Bias in Preconditioned Language Model Optimizers
GLiNER2-PII: A Multilingual Model for Personally Identifiable Information Extraction
GLiGuard: Schema-Conditioned Classification for LLM Content Moderation
Pioneer Agent: Continual Improvement of Small Language Models in Production
GLiNER: Generalist Model for Named Entity Recognition using Bidirectional Transformer
GLiNER2: An Efficient Multi-Task Information Extraction System with Schema-Driven Interface
Beyond Reactivity: Measuring Proactive Problem solving in LLM Agents