ToolOrchestra: Small Maestros, Big Intelligence

Kari Jaaskelainen

27 Nov 2025 — 1 min read

Small model, big wins

Large language models are great generalists, but really tough, multi-step problems still strain both brains and budgets. ToolOrchestra flips the script: instead of one giant model, a small “orchestrator” coordinates other models and specialized tools.

Trained with reinforcement learning that rewards outcomes, efficiency, and user preferences, the 8B-parameter Orchestrator chooses which tools to call, when, and how often—and it adapts to new tools it hasn’t seen before.

Results from the paper’s benchmarks: on Humanity’s Last Exam (HLE), Orchestrator scores 37.1%, topping GPT-5’s 35.1% while being 2.5× more efficient. On tau2-Bench and FRAMES, it surpasses GPT-5 by a wide margin at roughly 30% of the cost. Across metrics, it delivers one of the best performance–cost trade‑offs.

Takeaway: composing many smart tools under a lightweight conductor isn’t just cheaper—it’s smarter. This could be the practical path to scalable, tool‑augmented reasoning systems.

Paper: https://arxiv.org/abs/2511.21689v1

Register: https://www.AiFeta.com

AI ToolOrchestra LLM Orchestration ReinforcementLearning Efficiency AgenticAI Research HLE tau2Bench FRAMES

Beyond URLs: Metadata That Makes LLMs Train Faster

Smarter LLMs, Faster—thanks to metadata What if training a large language model didn't just rely on text, but on the context around it? This study shows that adding fine-grained metadata—not just URLs—can meaningfully speed up pretraining and improve quality. * Beyond URLs: detailed quality signals (e.

Escaping the Verifier: Learning to Reason via Demonstrations

LLMs can learn to reason—without task verifiers Many real-world problems don’t have automatic checkers to grade answers, even though we have lots of expert solutions. RARO (Relativistic Adversarial Reasoning Optimization) shows how to train reasoning skills from those examples alone. How it works: * A policy (the model) tries

VacuumVLA: A Two-in-One Robot Hand That Grips and Suctions

Robots guided by Vision-Language-Action (VLA) AI are getting better at everyday tasks—but most still use simple two-finger grippers. That limits them on smooth, flat, or handleless objects. VacuumVLA is a low-cost robot hand that merges a standard two-finger gripper with a vacuum suction cup. The robot can switch between

ChatDRex: No‑Code, Conversational AI for Drug Repurposing

Meet ChatDRex Finding new uses for approved drugs can save years and millions. But making those predictions usually takes teams of specialists and a tangle of tools. ChatDRex changes that. It’s a conversation-based, no‑code, multi‑agent system that lets clinicians and researchers ask complex bioinformatics questions in plain

Small model, big wins

Read more

Beyond URLs: Metadata That Makes LLMs Train Faster

Escaping the Verifier: Learning to Reason via Demonstrations

VacuumVLA: A Two-in-One Robot Hand That Grips and Suctions

ChatDRex: No‑Code, Conversational AI for Drug Repurposing