List your company

Policy-trajectory compliance evaluation on POLICYGUARDBENCH

88.83F1 ScoreLLAMA-3.3-70B-INSTRUCT
81.70683.555585.40587.2545Oct 3, 2025

Evaluation results

MethodLinks
LLAMA-3.3-70B-INSTRUCTSIZE=70B, LATENCY=305.0, FLOPS=45.732025.1088.832.9125
GEMMA-3-12B-ITSIZE=12B, LATENCY=51.3, FLOPS=7.872025.1087.7317.1014
POLICYGUARD-4BSIZE=4B, LATENCY=22.5, FLOPS=2.572025.1087.5938.9289
QWEN3-235B-A22B-INSTRUCT-2507SIZE=235B, LATENCY=3640.0, FLOPS=13.802025.1086.90.2387
CLAUDE-SONNET-4LATENCY=1238.02025.1086.780.701
QWEN2.5-72B-INSTRUCTSIZE=72B, LATENCY=205.0, FLOPS=48.022025.1086.074.1985
GEMMA-3-27B-ITSIZE=27B, LATENCY=73.6, FLOPS=17.992025.1085.211.5761
GEMINI-1.5-PROLATENCY=596.12025.1085.021.4263
DEEPSEEK-V3.1 (NON-THINKING)SIZE=685B, LATENCY=3270.02025.1084.070.2571
LLAMA-4-SCOUT-17B-16E-INSTRUCTSIZE=109B, LATENCY=265.0, FLOPS=63.922025.1081.983.0936