Articles from huggingface.co

NVIDIA argues that open synthetic data is essential for building inspectable, trustworthy AI agents while preserving proprietary secrets and fostering community collaboration.

Specialization is inevitable across optimization, biology, markets, and ML because fit beats breadth under finite resource constraints.

Gemma 4 E2B runs locally in-browser via WebGPU, letting users prompt the model directly without server-side inference.

Moonshot AI releases Kimi K2.7 Code, a 1T-parameter MoE coding model that improves real-world agentic tasks while cutting thinking tokens by 30% over K2.6.

Surya OCR 2 is a 650M-parameter vision-language model achieving 83.3% on olmOCR-bench and 87.2% on a 91-language benchmark, offering layout analysis, table recognition, and OCR in a single model.

AI evaluation costs now rival or exceed training costs, with agent benchmarks reaching $40,000 per sweep and training-in-the-loop evals resisting effective compression.

NVIDIA's Nemotron OCR v2 uses 12 million synthetic training images to achieve near-zero error rates across six languages at 34.7 pages per second on an A100.

Hugging Face built a Claude Code Skill and test harness that helps contributors port language models from transformers to mlx-lm while preserving code quality and reviewer trust.

ServiceNow's EVA framework jointly evaluates voice agents on task accuracy and conversational experience, revealing a consistent accuracy-experience tradeoff across 20 systems.

NVIDIA details a six-command pipeline using synthetic data generation and hard negative mining to fine-tune an embedding model on a single GPU in under a day, achieving over 10% retrieval improvement.

OpenEnv framework from Meta and Hugging Face reveals tool-using agents fail at multi-step reasoning, ambiguity resolution, and execution quality in realistic calendar environments.

Hugging Face released an open-source 'skill' that lets Claude Code, Codex, and Gemini CLI autonomously fine-tune LLMs on cloud GPUs and push models to the Hub.