AI2D
AI2D results reported for 3 models across 1 version. Updated 2026-10-02.
Report an issueVersions
1 version · Metric: Accuracy
| Version | Released | Models | Results |
|---|---|---|---|
| AI2D TESTLatest | March 24, 2016 | 3 | 3 |
Recently reported results: AI2D TEST
Latest version: AI2D TEST
| Model | Score | Source |
|---|---|---|
| Qwen3.6-35B-A3B | 92.7% | alibabacloud.com for Qwen3.6-35B-A3B on AI2D TEST (opens in a new tab)AccuracyReported Launch postAlibaba Cloud CommunitySelf-reportedQwen TeamRecord for Qwen3.6-35B-A3B on AI2D TEST |
| Qwen3.6-Plus | 94.4% | qwen.ai for Qwen3.6-Plus on AI2D TEST (opens in a new tab)AccuracyReported Launch postQwen TeamSelf-reportedRecord for Qwen3.6-Plus on AI2D TEST |
| Phi-4 Reasoning Vision 15B (automatic think/nothink) | 84.8% | huggingface.co for Phi-4 Reasoning Vision 15B (automatic think/nothink) on AI2D TEST (opens in a new tab)AccuracyReported Model cardMicrosoftSelf-reportedRecord for Phi-4 Reasoning Vision 15B on AI2D TEST |