List your company

Deontic Reasoning on DeonticBench SARA-Bin

90.6Macro F1Qwen3.5-397B
44.63256.56668.580.434Jun 3, 2026

Evaluation results

MethodLinks
Qwen3.5-397BHarness=terminus-kira2026.0690.6
Qwen3.5-397BHarness=claude-code2026.0688.9
Kimi-K2Harness=terminus-kira2026.0688.5
Qwen3.5-35BHarness=terminus-22026.0683.3
Qwen3.5-35BHarness=claude-code2026.0683.3
Qwen3.5-397BHarness=terminus-22026.0683.3
Qwen3.5-35BHarness=terminus-kira2026.0682.9
Qwen3.5-122BHarness=terminus-kira2026.0682.3
Qwen3-Coder-480BHarness=claude-code2026.0680
Qwen3.5-122BHarness=terminus-22026.0680
Qwen3.5-122BHarness=codex2026.0679.9
Qwen3-Coder-480BHarness=terminus-22026.0679.3
Qwen3.5-122BHarness=claude-code2026.0679.3
Qwen3.5-397BHarness=direct2026.0678.2
Qwen3-Coder-480BHarness=terminus-kira2026.0676.6
Qwen3.5-122BHarness=direct2026.0675.3
Qwen3.5-35BHarness=direct2026.0674
Kimi-K2Harness=codex2026.0673.3
Qwen3-235BHarness=codex2026.0672.1
Kimi-K2Harness=direct2026.0668.4
Qwen3-235BHarness=direct2026.0666.5
Qwen3-235BHarness=terminus-kira2026.0666.5
Qwen3-Coder-480BHarness=codex2026.0659.8
Qwen3-235BHarness=terminus-22026.0659.8
GPT-5.2Harness=direct2026.0659.7
Kimi-K2Harness=terminus-22026.0659.3
Qwen3-Coder-480BHarness=direct2026.0659.1
GPT-5.2Harness=terminus-kira2026.0656.9
GPT-5.2Harness=terminus-22026.0653.1
GPT-5.2Harness=codex2026.0646.4