List your company

Information Retrieval on LongSealQA

89.1RecallHarness-1
22.74839.97457.274.426Jun 1, 2026

Evaluation results

MethodLinks
Harness-1Parameters=20B, Category=Open-source small models2026.0689.189.194.4
Tongyi DeepResearchParameters=30B, Category=Open-source small models2026.0685.585.585.5
GPT-5.4Category=Frontier models2026.0682.482.488.5
Opus-4.6Category=Frontier models2026.0682.382.387.6
Sonnet-4.6Category=Frontier models2026.0678.378.384
Kimi-K2.5Category=Frontier models2026.0666.166.191.1
Context-1Parameters=20B, Category=Open-source small models2026.0656.356.385.8
Search-R1Parameters=32B, Category=Open-source small models2026.0651.451.451.4
GPT-OSSParameters=120B, Category=Frontier models2026.0646.446.491.4
Qwen3Parameters=32B, Category=Open-source small models2026.0642.742.770.4
GPT-OSSParameters=20B, Category=Open-source small models2026.0625.325.381.6