NVIDIA Nemotron 3 Ultra Leads Open Models on Accuracy and Efficiency in Agentic RTL Coding

Summary

Register transfer level (RTL) development relies heavily on iterative verification and tool feedback, which poses challenges for standard code generation models. To address realistic hardware design tasks, the ACE-RTL agent pairs an iterative generate-test-reflect workflow with NVIDIA Nemotron 3 Ultra. The model utilizes a hybrid Mamba-Attention Mixture-of-Experts architecture trained on a synthetic dataset encompassing code generation, editing, and error debugging. Across the comprehensive Verilog design problems benchmark, ACE-RTL with Nemotron 3 Ultra attained a 97.1% average pass rate, outperforming GLM 5.2 and Kimi K2.6. Additionally, Nemotron 3 Ultra achieved its accuracy while consuming an average of 6,629 tokens per iteration, which is substantially lower than competing models.

Takeaways

1. 1

   ACE-RTL with Nemotron 3 Ultra achieved a 97.1% average pass rate across nine CVDP task categories, outperforming Kimi K2.6 (95.2%) and GLM 5.2 (92.1%).

2. 2

   Nemotron 3 Ultra averaged 6,629 tokens per iteration on CVDP tasks, representing roughly 28% fewer tokens than GLM 5.2 (9,156) and 71% fewer than Kimi K2.6 (22,579).

3. 3

   Nemotron 3 Ultra uses a 550B total, 55B active parameter hybrid Mamba-Attention MoE architecture supporting a 1M-token context length.