Analyzes how on-policy distillation (OPD) transfers teacher LLM capabilities to student models across in-domain shifts, cross-domain transfer, and multi-teacher settings. Key findings: OPD conveys reasoning patterns rather than specific answers, same-origin teacher-student pairs generalize broadly, and multi-teacher combinations induce mixture-dependent tradeoffs.
Presents a unified black-box reinforcement learning framework to train and optimize agents running inside complex execution harnesses. Uses sandbox-parallel rollouts, a serving proxy that captures model calls and reconstructs multi-turn trajectories as prefix trees, and adapted GRPO/PPO optimizers to achieve stable, scalable RL across heterogeneous harnesses.
Injects proprietary news, regulatory and legal data into an open checkpoint via data-centric continual learning to improve performance on legal, tax and journalism tasks while preserving general capabilities and very long context support.
Evaluates whether AI systems can independently carry out project-level scientific research by progressively removing human methodological guidance across 60 tasks in 11 domains. Built with expert review, sandbox execution, and multi-agent–model scoring to measure innovation and autonomous experimental execution.
A 9B open-weight reasoning LLM that uses a self-improvement loop to auto-generate tasks, construct scaffolds, and optimize rollouts for stronger agentic coding and long-context reasoning. Single-GPU deployable, supports tool-calling and a 262,144-token context window.
Turns embodied navigation into 2D visual prompting where a vision-language model selects image pixels that are projected to 3D actions; adds selective chain-of-thought, compressed anchor-trajectory memory, and a two-level alignment objective to improve sample and runtime efficiency.
Fine-tunes long-horizon LLM agents with evolution strategies so full-model updates run at inference-level GPU memory. Emphasizes trajectory-level credit via black-box rewards, online prompt–parameter co-evolution, and a cosine decay for perturbation scale to balance exploration and adaptation; suited for limited-GPU settings.
A 35B mixture-of-experts LLM tuned for agentic coding and end-to-end self-improvement: it jointly generates tasks, scaffolds, and solution rollouts. Activates ~3B params/token, supports 256K context (extendable), and emits chain-of-thought plus OpenAI-style tool calls.
GGUF build of Ornith-1.5's 35B mixture-of-experts model (A3B) for local inference — activates ~3B params per token, supports up to 262,144 tokens, emits separate reasoning traces and OpenAI-style tool calls, optimized for agentic coding and long-context use cases.
A 9B dense reasoning LLM optimized for single‑GPU deployment and terminal-based coding agents, with long-context support (up to 262,144 tokens) and GGUF/quantized builds for edge/mobile. Strong on coding and agentic benchmarks.
Converts 200+ hours of expert Figma screen recordings into 3,469 Playwright-MCP action trajectories for training and evaluating vision-language and GUI agents; includes 126 long‑horizon tasks, phase labels, a 10‑skill taxonomy, and is CC‑BY‑4.0 licensed.
Provides 1,021.64 hours across 597 CAD/BIM workflows with synchronized screen recordings and interaction logs; each workflow includes video, timestamped input events, task specs, source files, final outputs, and evaluation rubrics for training or evaluating desktop CAD agents.