Mathematical Reasoning on AIME-120
68.02AccuracyReQAT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ReQATBit-Precision=W4A16 MXFP42026.06 | 68.02 | — | — | |
| ReQATBit-Precision=W4A4KV4 NVFP42026.06 | 65.63 | — | — | |
| FTBit-Precision=BF162026.06 | 64.79 | — | — | |
| BF16 BaselineModel=Qwen3-4B, W-Bits=BF162026.01 | 58.89 | 70.63 | — | |
| BF16Model=Qwen3-4B, W-Bits=BF162026.01 | 58.89 | — | — | |
| BaselineBit-Precision=BF162026.06 | 56.83 | — | — | |
| PTQBit-Precision=W4A16 MXFP42026.06 | 50.37 | — | — | |
| PTQBit-Precision=W4A4KV4 NVFP42026.06 | 50.13 | — | — | |
| Reasoning-QATModel=Qwen3-4B, W-Bits=W3G1282026.01 | 41.11 | — | — | |
| Reasoning-QATModel=Qwen3-4B, W-Bits=W4A4KV42026.01 | 36.67 | 60.78 | -9.85 | |
| GPTQModel=Qwen3-4B, W-Bits=W3G1282026.01 | 34.17 | — | — | |
| FlatQuantModel=Qwen3-4B, W-Bits=W4A4KV42026.01 | 32.78 | 58.28 | -12.35 | |
| AWQModel=Qwen3-4B, W-Bits=W3G1282026.01 | 25 | — | — | |
| Reasoning-QATModel=Qwen3-4B, W-Bits=W2G1282026.01 | 22.78 | — | — | |
| BF16 BaselineModel=R1-1.5B, W-Bits=BF162026.01 | 21.67 | 48.72 | — | |
| BF16Model=R1-Qwen-1.5B, W-Bits=BF162026.01 | 21.67 | — | — | |
| Reasoning-QATModel=R1-Qwen-1.5B, W-Bits=W3G1282026.01 | 16.39 | — | — | |
| Reasoning-QATModel=R1-1.5B, W-Bits=W4A4KV42026.01 | 12.5 | 41.31 | -7.41 | |
| BF16 BaselineModel=Qwen3-0.6B, W-Bits=BF162026.01 | 11.11 | 41.1 | — | |
| BF16Model=Qwen3-0.6B, W-Bits=BF162026.01 | 11.11 | — | — | |
| FlatQuantModel=R1-1.5B, W-Bits=W4A4KV42026.01 | 10 | 38.39 | -10.33 | |
| GPTQModel=R1-Qwen-1.5B, W-Bits=W3G1282026.01 | 10 | — | — | |
| Reasoning-QATModel=R1-Qwen-1.5B, W-Bits=W2G1282026.01 | 5.15 | — | — | |
| Reasoning-QATModel=Qwen3-0.6B, W-Bits=W3G1282026.01 | 3.89 | — | — | |
| AWQModel=R1-Qwen-1.5B, W-Bits=W3G1282026.01 | 3.33 | — | — | |
| RTNModel=R1-Qwen-1.5B, W-Bits=W3G1282026.01 | 0.83 | — | — | |
| Reasoning-QATModel=Qwen3-0.6B, W-Bits=W2G1282026.01 | 0.56 | — | — | |
| FlatQuantModel=Qwen3-0.6B, W-Bits=W4A4KV42026.01 | 0.28 | 17.34 | -23.76 | |
| QuaRotModel=Qwen3-0.6B, W-Bits=W4A4KV42026.01 | 0 | 4.84 | -36.26 | |
| Reasoning-QATModel=Qwen3-0.6B, W-Bits=W4A4KV42026.01 | 0 | 21.44 | -19.66 | |
| QuaRotModel=R1-1.5B, W-Bits=W4A4KV42026.01 | 0 | 2.11 | -46.61 | |
| RTNModel=Qwen3-0.6B, W-Bits=W3G1282026.01 | 0 | — | — | |
| GPTQModel=Qwen3-0.6B, W-Bits=W3G1282026.01 | 0 | — | — | |
| AWQModel=Qwen3-0.6B, W-Bits=W3G1282026.01 | 0 | — | — | |
| GPTQModel=Qwen3-0.6B, W-Bits=W2G1282026.01 | 0 | — | — | |
| AWQModel=Qwen3-0.6B, W-Bits=W2G1282026.01 | 0 | — | — | |
| GPTQModel=R1-Qwen-1.5B, W-Bits=W2G1282026.01 | 0 | — | — | |
| AWQModel=R1-Qwen-1.5B, W-Bits=W2G1282026.01 | 0 | — | — | |
| RTNModel=Qwen3-4B, W-Bits=W3G1282026.01 | 0 | — | — | |
| GPTQModel=Qwen3-4B, W-Bits=W2G1282026.01 | 0 | — | — | |
| AWQModel=Qwen3-4B, W-Bits=W2G1282026.01 | 0 | — | — |