AI agents are advancing quickly, but enterprise work remains a much harder test for agents.
Today coinciding with ICML, we are introducing FlowerBench: a frontier agent benchmark for evaluating AI agents on secure, proprietary, long-horizon enterprise tasks.
FlowerBench evaluates AI agents on real enterprise work while always keeping private data, tools, and context inside the organization. Evaluations are coordinated across the Flower Enterprise Evaluation Network, where only sanitized non-sensitive results are shared.
We have already assessed proprietary and open agents and models across a range of enterprise-grade tasks in various industries: Healthcare, Insurance, Operations, MLOps, Legal, Marketing and Finance.
Looking forward to feedback from people working on agent benchmarks, enterprise AI, tool-use evals, and long-horizon tasks evaluation.
Today coinciding with ICML, we are introducing FlowerBench: a frontier agent benchmark for evaluating AI agents on secure, proprietary, long-horizon enterprise tasks.
FlowerBench evaluates AI agents on real enterprise work while always keeping private data, tools, and context inside the organization. Evaluations are coordinated across the Flower Enterprise Evaluation Network, where only sanitized non-sensitive results are shared.
We have already assessed proprietary and open agents and models across a range of enterprise-grade tasks in various industries: Healthcare, Insurance, Operations, MLOps, Legal, Marketing and Finance.
Looking forward to feedback from people working on agent benchmarks, enterprise AI, tool-use evals, and long-horizon tasks evaluation.
*Other Useful Links*
Blogpost: https://flower.ai/blog/2026-07-06-flowerbench
Share your Agent Task: https://flowerlabs.typeform.com/to/AKGDGUhf