The last two years have seen NVIDIA's content safety stack grow from a focused English text classifier into a family of specialized models—each extending coverage to new modalities, languages, and…
Hugging Face redesigned its hf CLI to provide agent-optimized output, auto-detecting coding agents like Claude Code and Codex to render TSV format instead of human-friendly tables, reducing token…
The post describes using Direct Preference Optimization (DPO) on OCR models to reduce text degeneration, using rejection pairs from the model's own failures.
Hugging Face announced that Reachy Mini's conversation app can now use remote tools hosted in public Hugging Face Spaces, called over MCP.
Hcompany released the Holo3.1 family of computer-use models, improving robustness across environments (web, desktop, mobile), agent frameworks, and deployment targets.
JetBrains releases Mellum2, a 12B-parameter Mixture-of-Experts model trained from scratch on text and code, with 2.5B active parameters per token, under Apache 2.0 license.
IBM Research publishes a blog post on Hugging Face discussing the importance of agent logic for scalable enterprise AI adoption.