Text model

Orchestrator-8B

ToolOrchestra is NVIDIA’s method for training small “orchestrator” models that decide when and how to call tools and other LLMs. The main model, Orchestrator-8B, beats GPT-5 on hard agent benchmarks like Humanity’s Last Exam while using about 30% of the cost and running roughly 2.5× faster.

Type
Text
Weights
Open
Downloads
2.6k
Likes
598
Pipeline
text generation
Library
transformers

Model hub

nvidia/Nemotron-Orchestrator-8B

Repo
nvidia/Nemotron-Orchestrator-8B
Access
Public
Technical tags and languages
transformers safetensors qwen3 text-generation conversational base_model:Qwen/Qwen3-8B base_model:finetune:Qwen/Qwen3-8B eval-results text-generation-inference endpoints_compatible
Open on Hugging Face

Papers

1

More from NVIDIA

11

About NVIDIA

1