Models from EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning (arXiv:2608.06197)
AI & ML interests
Large Language Models, Tool Usage, Functional Calling
Recent Activity
Organization Card
ACE Team
Welcome to the official Hugging Face organization of ToolACE models and datasets from ACE Team.
In this organization, you can find our latest models and datasets. You can find:
Models
- ToolACE-8B: A fine-tuned model based on LLaMA-3.1-8B-Instruct tailed for functional calling, one of the best open-sourced models on BFCL leaderboard.
Datasets
- ToolACE: A functional calling dataset, aiming to enhance the functional-calling ability of LLMs. A subset of the training set of ToolACE-8B
models 5
Team-ACE/EnvACE-Qwen3-1.7B
Text Generation • 2B • Updated • 327
Team-ACE/EnvACE-Qwen3-8B
Text Generation • 8B • Updated • 314
Team-ACE/ToolACE-2.5-Llama-3.1-8B
8B • Updated • 15 • 4
Team-ACE/ToolACE-8B
8B • Updated • 8.31k • 77
Team-ACE/ToolACE-2-Llama-3.1-8B
8B • Updated • 256 • 52