FinHardBench - System Configuration: leaderboard
Metric: Mean best end-to-end latency in nanoseconds (lower is better; mean over 5 seeds of the lowest pipeline latency the model reaches in 24 feedback-driven rounds choosing among 432 configurations of a fixed 6-stage HFT FPGA pipeline under a 5,000-LUT budget; the optimum is 107.5 ns). Source: arxiv.org. Saturation forecast: Around June 2027. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 107.5 |
| 2 | GPT-5.4 | 107.5 |
| 3 | MiniMax-M2.7 | 108.5 |
| 4 | DeepSeek V3.2 | 116.1 |
| 5 | Gemini 3.1 Flash Lite (Preview) | 121.4 |
| 6 | Mistral Large 3 | 129.4 |
Interactive version: theaggregate.ai/benchmark?slug=finhardbench-system-configuration · How It Works · Data refreshed daily, snapshot 2026-09-29.