SimpleQA
AI benchmark results and model performance
Type: percentage
Recorded results
87
Average score
36.60%
Score range
1.80% - 97.10%
Leading model
97.10% - DeepSeek OCR 2
Scores over time
Individual benchmark scores plotted by date.
Model results
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| DeepSeek OCR 2 | - | 97.10% | inferred family alias from deepseek-v3.2-exp (score=0.3809; benches=14) | Yes | Source | |
| DeepSeek V3.2 Exp | 29 Sept 2025 | 97.10% | - | Yes | Source | |
| DeepSeek V3.1 | 21 Aug 2025 | 93.40% | Search agent evaluation | Yes | Source | |
| DeepSeek V3.1 Terminus | 22 Sept 2025 | 93.40% | inferred alias from deepseek-v3.1 | Yes | Source | |
| Ernie 5.0 | 22 Jan 2026 | 75% | - | Yes | Source | |
| Ernie 5.0 0110 | - | 75% | inferred version-family alias from ernie-5.0 | Yes | Source | |
| Ernie 5.0 Preview 1203 | - | 75% | inferred version-family alias from ernie-5.0 | Yes | Source | |
| Ernie 5.0 Preview 1220 | - | 75% | inferred version-family alias from ernie-5.0 | Yes | Source | |
| Gemini 3 Pro Image Preview (Nano Banana Pro) | 20 Nov 2025 | 72.10% | inferred modality/version alias from gemini-3-pro-preview | Yes | Source | |
| Gemini 3 Pro Preview | 18 Nov 2025 | 72.10% | - | Yes | Source | |
| Gemini 3 Flash Preview | 17 Dec 2025 | 68.70% | - | Yes | Source | |
| GPT 4.5 Preview | 27 Feb 2025 | 62.50% | - | Yes | Source | |
| Qwen 3 VL 32B Thinking | - | 55.40% | - | Yes | - | |
| Qwen 3 A235 A22B Instruct 2507 | - | 54.30% | - | Yes | Source | |
| Gemini 2.5 Pro Preview (2025-06-05) | 05 Jun 2025 | 54% | - | Yes | Source | |
| Gemini 2.5 Pro Preview TTS (2025-05-20) | 20 May 2025 | 54% | inferred family alias from gemini-2.5-pro-preview-06-05 (score=0.4243; benches=13) | Yes | Source | |
| Qwen 3 VL 235B A22B Instruct | - | 51.90% | - | Yes | - | |
| Gemini 2.5 Computer Use Preview | 07 Oct 2025 | 50.80% | inferred family alias from gemini-2.5-pro (score=0.3960; benches=16) | Yes | Source | |
| Gemini 2.5 Pro Experimental (2025-03-25) | 25 Mar 2025 | 50.80% | inferred alias from gemini-2.5-pro | Yes | Source | |
| Gemini 2.5 Pro Preview (2025-05-06) | 06 May 2025 | 50.80% | - | Yes | Source | |
| Gemini 2.5 Pro Preview TTS (2025-12-10) | 10 Dec 2025 | 50.80% | inferred modality/version alias from gemini-2.5-pro | Yes | Source | |
| Gemini Embedding 2 Preview | 10 Mar 2026 | 50.80% | manual fallback alias from gemini-2.5-pro | Yes | Source | |
| Qwen 3 8B | - | 49.60% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | Yes | - | |
| Qwen 3 Guard Stream 8B | - | 49.60% | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | Yes | - | |
| Qwen 3 VL 8B Thinking | - | 49.60% | - | Yes | - |
