Mathematical Reasoning on GAOKAO-MM Math
68.8AccuracyDS-R1-Distill-Qwen-32B + GeoTikzBridge-Base
Evaluation Results
| Method | Links | |
|---|---|---|
| DS-R1-Distill-Qwen-32B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 68.8 | |
| GLM-4.5-Air-106B-A12B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 68.8 | |
| GLM-Z1-32B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 67.5 | |
| GLM4.5-V-106B-A12B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 66.3 | |
| Skywork-OR1-32B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 66.3 | |
| InternVL3-78BTikz-supplementation=false2026.03 | 63.8 | |
| InternVL3.5-38B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 63.5 | |
| GPT-OSS-120B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 62.5 | |
| Qwen2.5-VL-72BTikz-supplementation=false2026.03 | 61.3 | |
| GLM4.5-V-106B-A12BTikz-supplementation=false2026.03 | 61.3 | |
| InternVL3.5-38BTikz-supplementation=false2026.03 | 60 | |
| Qwen3-VL-30B-A3B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 58.8 | |
| Qwen3-VL-30B-A3BTikz-supplementation=false2026.03 | 55 | |
| Qwen3-30B-A3B + GeoTikzBridge-BaseTikz-supplementation=true2026.03 | 55 | |
| GPT-4-turboTikz-supplementation=false2026.03 | 50 | |
| Qwen2.5-VL-32BTikz-supplementation=false2026.03 | 50 | |
| GPT-4VTikz-supplementation=false2026.03 | 45 |