List your company

General and STEM reasoning on MedQA (pass@k)

86.8Pass@1MINDLOOM
58.699265.994673.2980.5854May 20, 2026

Evaluation results

MethodLinks
MINDLOOMModel Family=Qwen3.5, Model Size=9B2026.0586.895.12
MegaScienceModel Family=Qwen3.5, Model Size=9B2026.0581.393.64
DS-V3.2 DistillModel Family=Qwen3.5, Model Size=9B2026.0579.9790.42
BaseModel Family=Qwen3.5, Model Size=9B2026.0578.7190.97
MINDLOOMModel Family=Qwen3.5, Model Size=4B2026.0578.2488.81
OpenThoughtModel Family=Qwen3.5, Model Size=9B2026.0577.3792.38
OpenThoughtModel Family=Qwen3.5, Model Size=4B2026.0575.0287.9
MegaScienceModel Family=Qwen3.5, Model Size=4B2026.0574.785.23
BaseModel Family=Qwen3.5, Model Size=4B2026.0574.6282.32
DS-V3.2 DistillModel Family=Qwen3.5, Model Size=4B2026.0573.0686.57
MINDLOOMModel Family=Qwen3-2507, Model Size=8B2026.0570.7786.8
DS-V3.2 DistillModel Family=Qwen3-2507, Model Size=8B2026.0567.4781.78
MegaScienceModel Family=Qwen3-2507, Model Size=8B2026.0567.3282.95
OpenThoughtModel Family=Qwen3-2507, Model Size=8B2026.0566.9281.07
MINDLOOMModel Family=Qwen3-2507, Model Size=4B2026.0565.1281.3
BaseModel Family=Qwen3-2507, Model Size=8B2026.056378.47
BaseModel Family=Qwen3-2507, Model Size=4B2026.0562.6974.7
MegaScienceModel Family=Qwen3-2507, Model Size=4B2026.0560.4877.92
DS-V3.2 DistillModel Family=Qwen3-2507, Model Size=4B2026.0560.1776.28
OpenThoughtModel Family=Qwen3-2507, Model Size=4B2026.0559.7874.94