Geometry problem solving on GeoQA
92.04Top-1 AccGemini-2.5-Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-2.5-ProDecoding=Greedy (τ=0), Normalization=Exact-match, Resolution=224x224, Prompting strategy=Identical prompts2026.02 | 92.04 | |
| GPT-5Decoding=Greedy (τ=0), Normalization=Exact-match, Resolution=224x224, Prompting strategy=Identical prompts2026.02 | 91.91 | |
| Qwen3-VLModel detail=Qwen3-VL-32B-Thinking, Decoding=Greedy (τ=0), Normalization=Exact-match, Resolution=224x224, Prompting strategy=Identical prompts2026.02 | 86.6 | |
| Qwen3-VL-7B-Instruct (GeoCode trained)Backbone=Qwen3-VL-7B-Instruct, Training Setting=Ours (GeoCode trained)2026.02 | 84.21 | |
| Qwen3-VL-7B-Instruct (Baseline)Backbone=Qwen3-VL-7B-Instruct, Training Setting=Baseline2026.02 | 83.16 | |
| GeoFocus-7BModel Category=Open-source Mathematical Geometry LMMs, Parameters=7B2026.02 | 71.9 | |
| Gemini 2.0 FlashModel Category=Closed-source LMMs2026.02 | 69.1 | |
| GeoUniModel Category=Open-source Mathematical Geometry LMMs2026.02 | 66.7 | |
| Qwen2.5-VL-7B-Instruct (GeoCode trained)Backbone=Qwen2.5-VL-7B-Instruct, Training Setting=Ours (GeoCode trained)2026.02 | 66.44 | |
| GeoGen-SFT-7BModel Category=Open-source Mathematical Geometry LMMs, Parameters=7B2026.02 | 64.6 | |
| GeoFocus-3BModel Category=Open-source Mathematical Geometry LMMs, Parameters=3B2026.02 | 64.3 | |
| Qwen2.5-VL-7B-Instruct (Baseline)Backbone=Qwen2.5-VL-7B-Instruct, Training Setting=Baseline2026.02 | 62.6 | |
| GeoGen-SFT-3BModel Category=Open-source Mathematical Geometry LMMs, Parameters=3B2026.02 | 61.4 | |
| Qwen2.5-VL-7BModel Category=Open-source LMMs, Parameters=7B2026.02 | 44.4 | |
| Gemma-3-4BModel Category=Open-source LMMs, Parameters=4B2026.02 | 42.6 | |
| InternVL-3-8BModel Category=Open-source LMMs, Parameters=8B2026.02 | 38.6 | |
| GPT-4oModel Category=Closed-source LMMs2026.02 | 34 | |
| Qwen2.5-VL-3BModel Category=Open-source LMMs, Parameters=3B2026.02 | 32.1 | |
| InternVL-3-2BModel Category=Open-source LMMs, Parameters=2B2026.02 | 31.2 | |
| InternVL-2.5-8BModel Category=Open-source LMMs, Parameters=8B2026.02 | 30.9 | |
| Claude 3.7Model Category=Closed-source LMMs2026.02 | 26.9 | |
| G-LLaVA-13BModel Category=Open-source Mathematical Geometry LMMs, Parameters=13B2026.02 | 20.1 | |
| mPLUG-Owl2-7BModel Category=Open-source LMMs, Parameters=7B2026.02 | 13 | |
| Deepseek-VL-7BModel Category=Open-source LMMs, Parameters=7B2026.02 | 12.2 | |
| Monkey-Chat-7BModel Category=Open-source LMMs, Parameters=7B2026.02 | 10.7 | |
| Chameleon-7BModel Category=Open-source LMMs, Parameters=7B2026.02 | 7.3 |