RAGEN-2 exposes a silent RL training failure where agents produce fluent but input-agnostic reasoning, invisible to entropy metrics, and fixes it with a lightweight reward-variance filter.
The command-line downloader shipped v1.32.9 with fresh extractors for six sites, a slew of fixes for Facebook, exhentai, and mangafire, plus new metadata fields.
Red Hat AI shrinks Google's 26B MoE Gemma 4 to FP8 with per-token dynamic quantization, cutting memory in half while keeping over 99% accuracy on most benchmarks.
IBM's new 8B dense model matches its previous 32B mixture-of-experts predecessor, ships under Apache 2.0 with a 131K context window and strong tool calling.
A free, open-source 6-module course teaching you to build a production-ready AI assistant on top of your personal knowledge base using agents, RAG, and fine-tuning.
DeltaWorld compresses each video frame into a single token, beating Cosmos with 35x fewer parameters and 2,000x fewer FLOPs while generating diverse futures in one pass