Meta AI
Open-Source LLM Entwickler (Llama)
Nougat
docprocessingVisual Transformer OCR (Meta) fuer wissenschaftliche Dokumente (Mathe, Formeln).
WebsiteMeta Llama 4
modelsLlama 4: Scout (10M Kontext!), Maverick, Behemoth. 200 Sprachen. Open Source.
Website<img alt="Self-Hosted LLM vs. Cloud APIs: How to Route Between Them" src="https://storage.ghost.io/c/0d/78/0d78b34c-0c5f...
arXiv:2610.28895v1 Announce Type: new Abstract: Cross-tokenizer logits distillation enables knowledge transfer across m...
arXiv:2610.29410v1 Announce Type: new Abstract: Preference optimization can improve every ranking metric while building...
arXiv:2505.24255v3 Announce Type: replace Abstract: Large Language Models (LLMs) have shown potential in simulating hum...
arXiv:2609.03966v2 Announce Type: replace Abstract: Agent evaluations report a tool-call rate read off the serving stac...
arXiv:2610.17093v2 Announce Type: replace Abstract: A model diff credits post-training with whatever separates a base c...
arXiv:2502.16435v5 Announce Type: replace-cross Abstract: Humans develop perception through a bottom-up hierarchy: from...
arXiv:2506.07121v3 Announce Type: replace-cross Abstract: Ensuring the safety and robustness of large language models (...
arXiv:2605.15341v2 Announce Type: replace-cross Abstract: Autonomous labs increasingly deploy large language models (LL...
arXiv:2605.26175v2 Announce Type: replace-cross Abstract: Low-bit activation quantization remains a major bottleneck in...