Davinci-Tool: Scaling Agentic Tool-Use Supervision with Verifiable Executable Environments

Key points

• Reliable agentic tool use requires training data that pair diverse multi-step workflows with executable evidence that the requested outcomes were achieved.1
• Davinci-ToolData preserves restorable states, execution traces, outcomes, and verifiers as a unified substrate for supervision.1
• We first train on these demonstrations with supervised fine-tuning, then reuse EES states and verifiers to compare alternative actions and construct outcome-grounded preferences for Direct Preference Optimization.1
• Davinci-Tool improves Qwen3-Coder-30B-A3B-Instruct from 52.4 to 88.2 on $τ^2$-bench, from 30.4 to 62.8 on BFCL v4 multi-turn, and from 16.9 to 56.2 on MCP-Atlas, outperforming Qwen3-Coder-480B-A35B-Instruct on all three benchmarks with only 1/16 of its total parameters.1

Sources (1)

• [1]Davinci-Tool: Scaling Agentic Tool-Use Supervision with Verifiable Executable Environments
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 12, 02:33 PM

  To address this challenge, we introduce Davinci-Tool, a 30B tool-use model, and Davinci-ToolData, an execution-grounded dataset designed to scale environment coverage and outcome verification together.

  Reliable agentic tool use requires training data that pair diverse multi-step workflows with executable evidence that the requested outcomes were achieved.

Extractive summary: sentences quoted from the sources.