2025-11-24

Models

• GitHub - Tencent-Hunyuan/HunyuanOCR
  • HunyuanOCR Usage Guide - vLLM Recipes
• black-forest-labs/FLUX.2-dev · Hugging Face
  • Diffusers welcomes FLUX-2
• GitHub - nari-labs/dia2: TTS model capable of streaming conversational audio in realtime.
• Zyphra
  • [2511.17127] Training Foundation Models on a Full-Stack AMD Platform: Compute, Networking, and System Design
• inclusionAI/LLaDA2.0-flash · Hugging Face
• microsoft/Fara-7B · Hugging Face
  • microsoft.com/en-us/research/wp-content/uploads/2025/11/Fara-7B-An-Efficient-Agentic-Model-for-Computer-Use.pdf

Papers

• [2511.08704] Rethinking generative image pretraining: How far are we from scaling up next-pixel prediction?
• [2511.19399] DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research
• [2511.18890] Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models
• [2511.16397] AICC: Parse HTML Finer, Make Models Better — A 7.3T AI-Ready Corpus Built by a Model-Based HTML Parser
• [2511.08704] Rethinking generative image pretraining: How far are we from scaling up next-pixel prediction?
• [2511.16249] Controllable Layer Decomposition for Reversible Multi-Layer Image Generation
• [2511.19365] DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation
  • GitHub - Zehong-Ma/DeCo: Official repository for “DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation”
• [2511.20635] iMontage: Unified, Versatile, Highly Dynamic Many-to-many Image Generation
• [2511.21631] Qwen3-VL Technical Report
• [2511.21691] Canvas-to-Image: Compositional Image Generation with Multimodal Controls

Code

• GitHub - opendatalab/MinerU: Transforms complex documents like PDFs into LLM-ready markdown/JSON for your Agentic workflows.
• GitHub - inclusionAI/dFactory: Easy and Efficient dLLM Fine-Tuning

Articles

• BFL - Representation Comparison
• FP8 Reinforcement Learning | Unsloth Documentation
• Continuous batching from first principles

Videos

• [ ]

Other

• [ ]

Tweets