Hi ACEBench Team,
We are excited to introduce D-CORE-8B, a new open-source model designed to improve multi-turn tool use in large reasoning models via our novel approach.
Resources:
📄 Paper: https://huggingface.co/papers/2602.02160
🤗 Model: https://huggingface.co/bowiehsu/D-CORE-8B
Technical Details:
We have verified the model locally against the latest BFCL v4 codebase.
Here is the implementation of the D-CORE handler:
https://github.com/alibaba/EfficientAI/tree/main/dcore/eval/acebench
We hope to collaborate with you to integrate and test this model, contributing to the broader agentic tool use community.
Thanks!
Hi ACEBench Team,
We are excited to introduce D-CORE-8B, a new open-source model designed to improve multi-turn tool use in large reasoning models via our novel approach.
Resources:
📄 Paper: https://huggingface.co/papers/2602.02160
🤗 Model: https://huggingface.co/bowiehsu/D-CORE-8B
Technical Details:
We have verified the model locally against the latest BFCL v4 codebase.
Here is the implementation of the D-CORE handler:
https://github.com/alibaba/EfficientAI/tree/main/dcore/eval/acebench
We hope to collaborate with you to integrate and test this model, contributing to the broader agentic tool use community.
Thanks!