List your company

Reasoning on HLE (test)

26AccuracyMTRouter
4.78410.29215.821.308Apr 26, 2026

Evaluation results

MethodLinks
MTRouterRouting Level=Multi-Turn Routers (turn-level)2026.042635
GPT-5Routing Level=Single-Model Baselines2026.0425.161.8
EmbedLLMRouting Level=Single-Turn Routers (episode-level)2026.0424.861.4
Router-R1Routing Level=Multi-Turn Routers (turn-level)2026.0424.251.9
LLM RouterRouting Level=Multi-Turn Routers (turn-level)2026.042456.2
AvengersProRouting Level=Single-Turn Routers (episode-level)2026.0423.747.5
Random RouterRouting Level=Multi-Turn Routers (turn-level)2026.042016.9
OpenRouterRouting Level=Multi-Turn Routers (turn-level)2026.0418.3138.5
DeepSeek-V3.2Routing Level=Single-Model Baselines2026.0415.622.4
RouterDCRouting Level=Single-Turn Routers (episode-level)2026.0412.810.3
Kimi-K2Routing Level=Single-Model Baselines2026.0411.412
GPT-OSS-120BRouting Level=Single-Model Baselines2026.049.70.7
MiniMax-M2Routing Level=Single-Model Baselines2026.047.818.1
Gemini-2.5-Flash-LiteRouting Level=Single-Model Baselines2026.045.63