Tag
Explore by tags
Cloud-native control plane that scales vLLM on Kubernetes, adding the routing, autoscaling, and fault tolerance single-instance serving lacks. Brings high-density LoRA management, an LLM gateway, distributed KV cache reuse, and SLO-aware GPU serving.
Open-source HybridFlow implementation for RL post-training of LLMs. Decouples control flow from compute so PPO, GRPO, GSPO and DAPO share one dataflow; pairs FSDP/Megatron with vLLM/SGLang rollout and reports 1.5-20x throughput over prior RLHF stacks.
High-accuracy biomolecular structure prediction suite: open-source models (protenix-v2/v1), a benchmark/evaluation toolkit, and a web server for inference. Targets protein/antibody–antigen and ligand-aware predictions with inference-time sampling and constraint support.
Drives your computer from natural language: a vision-language model reads raw screenshots and works the mouse and keyboard like a person, controlling any GUI app without APIs or accessibility hooks. Local or remote operator modes on Windows and macOS.
Orchestrates a lead agent, isolated parallel sub-agents, long-term memory, and sandboxes for long-horizon tasks — minutes to hours of deep research, coding, and content creation. LangChain/LangGraph-based with extensible skills; v2 is a full rewrite.
Converts document images—scans, photos, born-digital PDFs—into structured text in two stages: first map layout and reading order, then parse each element (text, tables, formulas, figures) in parallel, each guided by its own task prompt.
Continuously screenshots your screen, feeds the captures to a vision-language model, then pushes back daily summaries, weekly recaps, and todos on its own. Local-first desktop app: data stays on your machine; runs on OpenAI-format or local LLMs.
Models an AI agent's context as a file system, unifying memory, resources, and skills instead of flat vector RAG. Uses L0/L1/L2 tiered loading to cut tokens, directory-recursive plus semantic retrieval, and visualized retrieval traces for debugging.
Delivers image and video generation, editing, and understanding inside a single 3B-parameter multimodal model trained from scratch with a multi-task recipe. Notable for strong unified benchmarks at 3B scale; inference requires large GPU memory (≈40GB+ VRAM).
Provides the renderer weights and inference code for Bernini’s video renderer, enabling text→video, image→video and video editing inference. Offers a ready diffusers-format bundle or safetensors checkpoints under Apache‑2.0; intended for multi‑GPU/Hopper inference and reproducible research.