Mathematics on AIME 25 (Req/s, Speedup, Acc.)
39.43Req/sSQUEEZE EVOLVE (p=0)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SQUEEZE EVOLVE (p=0)Model 1=GPT-OSS-20B, Model 2=GPT-OSS-120B, Obs. M1 share=100%, GPU split (G2:G1)=0:202026.04 | 39.43 | 2.31 | 90.8 | |
| SQUEEZE EVOLVE (p=10)Model 1=GPT-OSS-20B, Model 2=GPT-OSS-120B, Obs. M1 share=87.5%, GPU split (G2:G1)=4:162026.04 | 24.59 | 1.44 | 90.5 | |
| RSAModel 1=GPT-OSS-20B, Model 2=GPT-OSS-120B, Obs. M1 share=0%, GPU split (G2:G1)=20:02026.04 | 17.09 | 1 | 90.1 | |
| SQUEEZE EVOLVE (p=0)Model 1=Qwen3-30B-A3B-I, Model 2=Qwen3-235B-A22B-I, Obs. M1 share=100%, GPU split (G2:G1)=0:162026.04 | 13.47 | 9.9 | 81 | |
| SQUEEZE EVOLVE (p=10)Model 1=Qwen3-30B-A3B-I, Model 2=Qwen3-235B-A22B-I, Obs. M1 share=88.9%, GPU split (G2:G1)=8:82026.04 | 7.41 | 5.44 | 80.1 | |
| RSAModel 1=Qwen3-30B-A3B-I, Model 2=Qwen3-235B-A22B-I, Obs. M1 share=0%, GPU split (G2:G1)=16:02026.04 | 1.36 | 1 | 82 |