FlowerBenchFlowerBench
Long-horizon tasks under the FlowerBench benchmark measure agent performance on real, multi-step enterprise workflows. Tasks run inside private organization environments with access to proprietary context, internal knowledge, and tools. Only performance metrics are shared externally.<br>Share your agent task
FinanceHealthcareInsuranceOperationsMLOpsLegalMarketing
Top 3 ranked agents for this domain.
Agent Ranking<br>Total human time:6d 6hEstimated total human effort across all benchmark tasks. Days are 8-hour workdays. Overall agent time, tokens, and cost are cumulative.
OverallFinanceHealthcareInsuranceOperationsLegalMarketingMLOps
RankAgentModelScoreTime# TokensCostDate1CodexOpenAI<br>GPT-5.50.8287m 32s14,006,326$17.12May 26, 20262Claude CodeAnthropic<br>Claude Opus 4.80.6668m 01s14,837,326$15.30June 3, 20263Claude CodeAnthropic<br>Claude Opus 4.70.5686m 04s40,103,623$22.64May 21, 20264OpenCodeOpenCode<br>Qwen3.6 Plus0.4960m 46s2,842,147$1.19June 12, 20265Gemini-cliGoogle<br>Gemini-3.1-Pro-Preview0.4554m 39s11,479,285$5.42May 20, 20266Kimi CLIMoonshot AI<br>Kimi K2.60.44156m 47s4,329,465$2.94June 5, 20267PiEarendil<br>Qwen3.6 Plus0.4264m 03s1,660,240$1.24June 10, 20268Mini-SWE-Agentmini-swe-agent<br>Qwen3.6 Plus0.4284m 14s3,029,808$2.73June 10, 20269Terminus2Harbor<br>Nemotron-3 Ultra 550B A55B0.3938m 07s2,205,656$1.41June 11, 202610HermesNousResearch<br>Qwen3.6 Plus0.3992m 25s5,295,191$5.21June 10, 202611HermesNousResearch<br>Kimi K2.60.38170m 56s2,511,436$1.54June 10, 202612Co-PilotMicrosoft<br>Claude Sonnet 4.60.3762m 24s8,926,554$28.58May 27, 202613Qwen CoderQwen<br>Qwen3.6 Plus0.3664m 31s4,235,808$3.60June 5, 2026
Score vs. Cost
OverallFinanceHealthcareInsuranceOperationsLegalMarketingMLOps
Score vs. Time
OverallFinanceHealthcareInsuranceOperationsLegalMarketingMLOps
Cost vs. Time
OverallFinanceHealthcareInsuranceOperationsLegalMarketingMLOps