AI Train2025
Trains multi-step LLM agents with reinforcement learning (GRPO) on your own tasks, wrapping existing agent code behind an OpenAI-compatible client. Its RULER mode scores trajectories with an LLM judge, so there's no reward function to hand-write.
AI Others2025
Argues AI has entered its 'second half': a working recipe (language pre-training priors + scale + reasoning) now generalizes RL across tasks, so the bottleneck shifts from inventing methods to defining problems and rethinking evaluation.
AI Train2025
Trains and optimizes AI agents with reinforcement learning using almost zero code change. Works with any agent framework (LangChain, OpenAI Agents SDK, AutoGen, CrewAI) or none, and can selectively optimize a single agent inside a multi-agent system.
Embodied AI2025
Provides a unified Python interface to collect data, train visual/dynamics world models, and evaluate them with model-predictive control across many standardized environments. Includes reference baselines, planning solvers, dataset converters, and LanceDB-backed formats for reproducible experiments. Best suited for researchers benchmarking world-model algorithms.
AI Infra2025
Provides Gymnasium-style APIs and tooling to run isolated, networked execution environments for agentic reinforcement learning. Offers async/sync EnvClients, Docker/Kubernetes container providers, a web UI and CLI for scaffolding and deploying environments (Hugging Face Spaces); experimental and evolving.
AI Infra2025
Provides a Gymnasium-style API and tooling to create, deploy, and interact with isolated execution environments for agentic RL training. Includes async/sync clients, a web interface, CLI, Docker-based deployment, and Hugging Face Spaces integration.
AI Train2025
Worked examples and reusable abstractions for fine-tuning open LLMs via the Tinker training API: you write the training loop while distributed execution runs remotely. Covers SFT, math/code RL, DPO, three-stage RLHF, distillation, and tool use.
Embodied AI2025
Contains training, evaluation, and deployment code plus checkpoints for humanoid whole-body controllers (Decoupled WBC and GEAR‑SONIC). Includes C++ inference, VR teleoperation, data pipelines (Bones‑SEED) and Hugging Face checkpoints for research-to-robot workflows.
An open large language model pairing DeepSeek Sparse Attention (DSA) for cheaper long-context inference with a scaled RL pipeline. Authors claim parity with GPT-5, with a high-compute Speciale variant surpassing it and rivaling Gemini-3.0-Pro on reasoning.
AI Dataset2025
Large-scale, real-world dual-arm video corpus for embodied robotics and reinforcement-learning research — over 1TB of multimodal recordings on Hugging Face, intended for training and evaluating agents in real manipulation scenarios; CC BY‑SA 4.0.
Embodied AI2026
Train robot reinforcement-learning agents with a heterogeneous runtime that streams CPU-parallel physics simulations (MuJoCo / Motrix) via shared memory into GPU/accelerator policy learners; provides a unified CLI, cross-platform backend support and demo checkpoints.
AI Model2026
Instruction-tuned 13B LLM post-trained on 260B tokens of pre-1931 English and finetuned with online DPO (LLM-as-judge) to improve instruction-following; suited for period-style English generation and etiquette/letter-writing formats, but not optimized for contemporary factual updates.