List your company

Hardware Vulnerability Detection on RTL Hardware Vulnerability

100AccuracyGemini 2.5 Flash
3.425628.497853.5778.6422Mar 5, 2026

Evaluation results

MethodLinks
Gemini 2.5 FlashFramework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0310042.86
Gemini 2.5 ProFramework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0310050
Avg. for the SoTA LLMsFramework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0390.4840.48
Gemini 2.0 FlashFramework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0385.7135.71
Gemini 2.5 Flash LiteFramework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0385.7150
GPT-4oFramework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0385.7135.71
GPT-5Framework=SecureRAG-RTL, Model Scale=SoTA LLMs2026.0385.7128.57
Qwen 3 8BFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.0371.4335.71
Gemma 3 4B ItFramework=SecureRAG-RTL, Model Scale=small LLMs2026.0364.2942.86
LLaMA 3.2 11B V-ItFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.0364.2928.57
Gemini 2.5 FlashFramework=Baseline, Model Scale=SoTA LLMs2026.0357.14—
GPT-5Framework=Baseline, Model Scale=SoTA LLMs2026.0357.14—
Avg. for medium LLMsFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.0352.3822.62
Qwen 2.5 Coder 3B ItFramework=SecureRAG-RTL, Model Scale=small LLMs2026.035028.57
Code Llama 7B ItFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.035021.43
Gemini 2.0 FlashFramework=Baseline, Model Scale=SoTA LLMs2026.0350—
Gemini 2.5 ProFramework=Baseline, Model Scale=SoTA LLMs2026.0350—
GPT-4oFramework=Baseline, Model Scale=SoTA LLMs2026.0350—
Avg. for the SoTA LLMsFramework=Baseline, Model Scale=SoTA LLMs2026.0350—
Falcon 3 3B ItFramework=SecureRAG-RTL, Model Scale=small LLMs2026.0342.8628.57
Granite 4.0 Micro 3BFramework=SecureRAG-RTL, Model Scale=small LLMs2026.0342.8628.57
StarCoder 2 7BFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.0342.8614.29
DeepSeek-Coder 7B ItFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.0342.8614.29
Mistral v0.3 7BFramework=SecureRAG-RTL, Model Scale=medium LLMs2026.0342.8621.43
Avg. for small LLMsFramework=SecureRAG-RTL, Model Scale=small LLMs2026.0340.4826.19
Qwen 3 8BFramework=Baseline, Model Scale=medium LLMs2026.0335.71—
LLaMA 3.2 11B V-ItFramework=Baseline, Model Scale=medium LLMs2026.0335.71—
Gemini 2.5 Flash LiteFramework=Baseline, Model Scale=SoTA LLMs2026.0335.71—
Avg. for medium LLMsFramework=Baseline, Model Scale=medium LLMs2026.0329.76—
Code Llama 7B ItFramework=Baseline, Model Scale=medium LLMs2026.0328.57—
StarCoder 2 7BFramework=Baseline, Model Scale=medium LLMs2026.0328.57—
DeepSeek-Coder 7B ItFramework=Baseline, Model Scale=medium LLMs2026.0328.57—
DeepSeek-R1-D-Qwen 1.5BFramework=SecureRAG-RTL, Model Scale=small LLMs2026.0321.4314.29
Qwen 2.5 Coder 3B ItFramework=Baseline, Model Scale=small LLMs2026.0321.43—
Phi 3 Mini 3.8B ItFramework=SecureRAG-RTL, Model Scale=small LLMs2026.0321.4314.29
Gemma 3 4B ItFramework=Baseline, Model Scale=small LLMs2026.0321.43—
Mistral v0.3 7BFramework=Baseline, Model Scale=medium LLMs2026.0321.43—
Falcon 3 3B ItFramework=Baseline, Model Scale=small LLMs2026.0314.29—
Granite 4.0 Micro 3BFramework=Baseline, Model Scale=small LLMs2026.0314.29—
Avg. for small LLMsFramework=Baseline, Model Scale=small LLMs2026.0314.29—
DeepSeek-R1-D-Qwen 1.5BFramework=Baseline, Model Scale=small LLMs2026.037.14—
Phi 3 Mini 3.8B ItFramework=Baseline, Model Scale=small LLMs2026.037.14—