Share your thoughts, 1 month free Claude Pro on usSee more

Visual Question Answering on SimpleVQA (Accuracy)

0.737Accuracy

Gemini-3-Flash

Updated 2d ago

Evaluation Results

Method	Links
Gemini-3-Flash 2026.04		0.737
SenseNova-MARS-8B 2026.04		0.702
POINTS-Seeker-8B 2026.04		0.688
Gemini-2.5 Flash 2026.04		0.686
MM-DeepResearch 32B 2026.03		0.676
GoG-3-8B-Think-RL 2026.01		0.6644
MM-DeepResearch-8B 2026.03		0.659
MM-DeepResearch-8B 2026.04		0.659
GoG-3-8B-Think-SFT 2026.01		0.6565
MIA 2026.04		0.649
GoG-2.5-7B-RL 2026.01		0.6486
GPT-4o 2026.03		0.636
Unsupervised MIA 2026.04		0.633
No Memory 2026.04		0.63
Qwen3-VL-8B-Think 2026.01		0.6269
ExpeL 2026.04		0.625
Qwen3-VL-8B 2026.03		0.623
MM-DeepResearch-7B 2026.03		0.62
Qwen3-VL-8B-Thinking 2026.01		0.619
Memento 2026.04		0.619
SenseNova-MARS-8B 2026.03		0.617
GPT-5 2026.04		0.617
GPT-4o 2026.04		0.616
ERNIE 5.0-Base 2026.02		0.6154
GoG-2.5-7B-SFT 2026.01		0.6061
RAG 2026.04		0.605
ReasoningBank 2026.04		0.604
DeepEyes-v2-7B 2026.03		0.594
DeepEyesV2 2025.11		0.594
Deepeyes2 2026.04		0.594
DeepEyesV2 2026.04		0.594
Claude-3.7-Sonnet 2026.04		0.593
Qwen2.5-VL-7B 2026.01		0.5913
WebWatcher-32B 2026.03		0.59
WebWatcher-32B 2026.04		0.59
Qwen3-VL-32B 2026.03		0.587
Qwen-2.5-VL-72B 2026.04		0.586
Qwen3-VL-32B 2026.03		0.58
MMSearch-R1-7B 2026.03		0.574
MMSearch-R1 2025.11		0.574
MMSearch-R1-7B 2026.04		0.574
Mem0 2026.04		0.567
GPT-5 2026.03		0.559
DeepMMSearch-R1 2026.04		0.559
DeepMMSearch-R1-7B 2026.04		0.559
DeepMMSearch-R1-7B 2026.03		0.558
GPT-5.4 2026.04		0.555
Gemini-2.5 Flash 2026.04		0.555
MMSearch-R1 2026.04		0.553
MMSearch-R1* 2026.01		0.5479
WebWatcher-7B 2026.03		0.543
WebWatcher 2025.11		0.543
WebWatcher 2026.04		0.543
WebWatcher-7B 2026.04		0.543
GPT-5 2026.03		0.541
Gemini 2.5 Pro 2025.11		0.534
Gemini-2.5-Pro 2026.04		0.534
Qwen3-VL-30B A3B-Instruct 2025.12		0.527
Gemini-2.5 flash-lite 2025.12		0.522
Qwen3-VL-8B 2026.03		0.52
GPT-4o 2026.03		0.517
Qwen-2.5-VL 2025.11		0.516
A-Mem 2026.04		0.516
Step-GUI-8B 2025.12		0.508
Qwen3-VL 8B-Instruct 2025.12		0.502
Qwen2.5-VL-32B+ReACT 2026.04		0.485
Step-GUI-4B 2025.12		0.481
Qwen3-VL-8B 2026.03		0.471
GPT4o 2025.11		0.466
GPT-4o 2026.04		0.466
Qwen3-VL-30B-A3B-Thinking 2026.01		0.4373
GPT-4o 2026.01		0.4344
Qwen3-VL-8B-Instruct 2026.04		0.429
Qwen2.5-VL-32B 2026.01		0.4284
Visual-ARFT-7B 2026.03		0.424
Qwen3-VL-30B-A3B-Instruct 2026.01		0.4235
Qwen3-VL-8B-Thinking 2026.01		0.4176
Qwen2.5-VL-7B 2026.01		0.4166
Qwen2.5-VL-7B 2026.01		0.4126
Qwen3-VL-8B-Instruct 2026.01		0.4018
Qwen2.5-VL-32B 2026.04		0.401
ACPO 2026.03		0.4
DPO-Shift 2026.03		0.397
SimPO 2026.03		0.395
GPT5-nano minimal 2025.12		0.39
β-DPO 2026.03		0.39
Instruct 2026.03		0.387
Qwen-2.5-VL 2025.11		0.384
DPO 2026.03		0.38
IPO 2026.03		0.38
ACPO 2026.03		0.38
β-DPO 2026.03		0.376
DPO-Shift 2026.03		0.375
IPO 2026.03		0.363
DPO 2026.03		0.358
Qwen2.5-VL-7B+ReACT 2026.04		0.358
SimPO 2026.03		0.355
Instruct 2026.03		0.344
Qwen2.5-VL-7B 2026.04		0.304