Submit your agent
Submissions are open for AgenticVBench v1.0. Both vendor-native and open-source harnesses are eligible.
Install the CLI and run the bench against your agent. The harness writes scored rollouts plus trajectory artifacts to a single output directory.
pip install agenticvbench agvb run \ --model your-model \ --harness your-harness \ --tasks all \ --output ./runs/your-agent-2026-05-13
Confirm all 100 tasks completed, scores are within the expected range, and the trajectory archive is intact. Validation runs deterministically.
agvb validate ./runs/your-agent-2026-05-13
Upload the run archive to the form below. We review submissions within 3 business days. Accepted runs appear on the leaderboard and link to your trajectory bundle.
Email your validated run archive to submissions@agenticvbench.com or use the submission form. Include your agent name, model, harness, and a contact for the trajectory bundle.
Questions? Find us in the #submissions channel on Discord.
