RealWorldQA
AI benchmark results and model performance
Type: percentageSpatial Reasoning
Recorded results
43
Average score
75.26%
Score range
64.20% - 86%
Leading model
86% - Seed 2.0 Pro
Scores over time
Individual benchmark scores plotted by date.
Model results
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| Seed 2.0 Pro | 14 Feb 2026 | 86% | Seed2 official benchmark table | RealWorldQA | Yes | Source | |
| Qwen 3.8 27B | 14 Aug 2026 | 85.90% | - | Yes | Source | |
| Qwen 3.6 Plus | 01 Apr 2026 | 85.40% | - | Yes | Source | |
| Qwen 3.5 122B A10B | 24 Feb 2026 | 85.10% | - | Yes | Source | |
| Qwen 3.5 35B A3B | 24 Feb 2026 | 84.10% | - | Yes | Source | |
| Qwen 3.5 27B | 24 Feb 2026 | 83.70% | - | Yes | Source | |
| Qwen 3.5 Flash | 23 Feb 2026 | 83.70% | inferred family alias from qwen3.5-27b (score=0.4147; benches=81) | Yes | Source | |
| Seed 2.0 Lite | 14 Feb 2026 | 81.70% | Seed2 official benchmark table | RealWorldQA | Yes | Source | |
| Seed 2.0 Mini | 14 Feb 2026 | 81.60% | Seed2 official benchmark table | RealWorldQA | Yes | Source | |
| Qwen 3 VL 235B A22B Thinking | - | 81.30% | - | Yes | - | |
| Qwen 3 VL 235B A22B Instruct | - | 79.30% | - | Yes | - | |
| Qwen 3 VL 32B Instruct | - | 79% | - | Yes | - | |
| Qwen 3 VL 32B Thinking | - | 78.40% | - | Yes | - | |
| Qwen 2 VL 2B | - | 77.80% | inferred family alias from qwen2-vl-72b (score=0.4083; benches=15) | Yes | Source | |
| Qwen 2 VL 72B | - | 77.80% | - | Yes | Source | |
| Qwen 3 VL 30B A3B Thinking | - | 77.40% | - | Yes | - | |
| Qwen 3 VL 30B A3B Instruct | - | 73.70% | - | Yes | - | |
| Qwen 3 8B | - | 73.50% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | Yes | - | |
| Qwen 3 Guard Stream 8B | - | 73.50% | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | Yes | - | |
| Qwen 3 VL 8B Thinking | - | 73.50% | - | Yes | - | |
| Qwen 3 VL Embedding 8B | - | 73.50% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50) | Yes | - | |
| Qwen 3 VL Reranker 8B | - | 73.50% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50) | Yes | - | |
| Qwen 3 4B Instruct 2507 | - | 73.20% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 4B SafeRL | - | 73.20% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 4B Thinking 2507 | - | 73.20% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - |
