List your company

Reasoning evaluation on 109-sample (test)

97.25AccuracyUniverse Routing
87.588490.096792.60595.1133Mar 16, 2026

Evaluation results

MethodLinks
Universe RoutingParams=465M2026.0397.25161—
Qwen3-NextParams=80B2026.0396.333,6402281
GLM-4.7Params=696B2026.0395.2812,3927750.131
Kimi-K2.5Params=1T2026.0394.59,8756170.371
Cogito-2.1Params=671B2026.0394.441,413880.289
GPT-OSSParams=120B2026.0391.741,9861240.077
DeepSeek-v3.1Params=671B2026.0387.964,0902560.01