List your company

Long-context reasoning on OOLONG

68.4Accuracyλ-RLM
8.49624.04839.655.152Mar 20, 2026Mar 22, 2026Mar 25, 2026Mar 28, 2026Mar 31, 2026Apr 3, 2026Apr 6, 2026

Evaluation results

MethodLinks
λ-RLMParadigm=P3, Model Family=Qwen3, Model Scale=235B2026.0368.4
RLMParadigm=P2, Model Family=Llama, Model Scale=405B2026.0363.8
λ-RLMParadigm=P3, Model Family=Qwen3, Model Scale=32B2026.0362.5
λ-RLMParadigm=P3, Model Family=Llama, Model Scale=405B2026.0361.7
λ-RLMParadigm=P3, Model Family=Llama, Model Scale=70B2026.0361.2
RLMParadigm=P2, Model Family=Qwen3, Model Scale=235B2026.0356.5
λ-RLMParadigm=P3, Model Family=Mistral, Model Scale=8x22B2026.0355.8
λ-RLMParadigm=P3, Model Family=Qwen3, Model Scale=8B2026.0348.3
RLMParadigm=P2, Model Family=Mistral, Model Scale=Cdsrl2026.0348.2
DirectParadigm=P1, Model Family=Llama, Model Scale=405B2026.0347.1
λ-RLMParadigm=P3, Model Family=Llama, Model Scale=8B2026.0345.7
λ-RLMParadigm=P3, Model Family=Mistral, Model Scale=Cdsrl2026.0345.6
RLMParadigm=P2, Model Family=Llama, Model Scale=70B2026.0345.3
DirectParadigm=P1, Model Family=Qwen3, Model Scale=235B2026.0344
RLMParadigm=P2, Model Family=Qwen3, Model Scale=32B2026.0342.7
GEMINI 3.1 FLASH-LITE2026.0442.38
λ-RLMParadigm=P3, Model Family=Mistral, Model Scale=7B2026.0340.2
QWEN3.5-35B-A3B-FP8Precision=FP82026.0439.76
RLMParadigm=P2, Model Family=Mistral, Model Scale=8x22B2026.0338.9
GPT-OSS-120B2026.0438.08
GEMINI-2.5-PRO2026.0436.56
GPT-OSS-20BReasoning Effort=High, Training Status=base2026.0435.86
QWEN3-4B-INSTRUCT-2507Training Status=SFT (π²-4B)2026.0435.48
DirectParadigm=P1, Model Family=Llama, Model Scale=70B2026.0334.8
GPT-OSS-20BReasoning Effort=Low, Training Status=SFT (π²-20B-A3B)2026.0433.31
GPT-OSS-20BReasoning Effort=High, Training Status=SFT (π²-20B-A3B)2026.0431.95
DirectParadigm=P1, Model Family=Qwen3, Model Scale=32B2026.0331.2
GPT-OSS-20BReasoning Effort=Low, Training Status=base2026.0431.03
QWEN3-4B-INSTRUCT-2507Training Status=base2026.0429.67
DirectParadigm=P1, Model Family=Mistral, Model Scale=8x22B2026.0328.5
RLMParadigm=P2, Model Family=Qwen3, Model Scale=8B2026.0324.1
DirectParadigm=P1, Model Family=Mistral, Model Scale=Cdsrl2026.0322.6
RLMParadigm=P2, Model Family=Llama, Model Scale=8B2026.0322.6
RLMParadigm=P2, Model Family=Mistral, Model Scale=7B2026.0318.3
DirectParadigm=P1, Model Family=Llama, Model Scale=8B2026.0314.3
DirectParadigm=P1, Model Family=Qwen3, Model Scale=8B2026.0312.5
DirectParadigm=P1, Model Family=Mistral, Model Scale=7B2026.0310.8