List your company
ResearchFormerly WizWand

Video Question Answering on Video-MME

86.9AccuracyGemini-3.0-Flash
6.809627.602348.39569.1877Jan 30, 2026Feb 24, 2026Mar 22, 2026Apr 16, 2026May 12, 2026Jun 6, 2026Jul 2, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Gemini-3.0-FlashZero-shot evaluation=true2026.0186.9
Gemini-2.5-ProZero-shot evaluation=true2026.0185.2
Qwen3-VL-8B + GRPOParams=8B, Training Scale=5.3K QA, Zero-shot evaluation=true2026.0176.78
Full PrecisionModel=Qwen2.5-VL-72B-Instruct, Weight bits=16 bit, Memory=136.74 GB, Evaluation protocol=zero-shot2026.0773.3
Qwen3-VL-8B (baseline)Params=8B, Training Scale=1.0T tokens2026.0171.4
Full PrecisionModel=Qwen2.5-VL-32B-Instruct, Weight bits=16 bit, Memory=62.31 GB, Evaluation protocol=zero-shot2026.0770.5
Qwen2-VL-7B + GRPOParams=7B, Training Scale=5.3K QA, Zero-shot evaluation=true2026.0169.12
Qwen2-VL-7B (baseline)Params=7B, Training Scale=1.2T tokens2026.0167.5
InternVideo2.5-8BParams=8B, Training Scale=16M clips2026.0165.4
GPTQModel=Qwen2.5-VL-72B-Instruct, Weight bits=3 bit, Memory=28.61 GB, Evaluation protocol=zero-shot2026.0765.11
SAB-LVLMModel=Qwen2.5-VL-72B-Instruct, Weight bits=1.07 bit, Memory=23.23 GB, Evaluation protocol=zero-shot2026.0764.93
HINT InternVL3-8BNumber of sampled frames=32, Backbone=InternVL3-8B2026.0364.6
InternVL3-8BNumber of sampled frames=322026.0364.2
LLaVA-Video-7BParams=7B, Training Scale=1.3M videos2026.0163.3
Qwen2-VL-2B + GRPOParams=2B, Training Scale=5.3K QA, Zero-shot evaluation=true2026.0162.11
Full PrecisionModel=Qwen2.5-VL-7B-Instruct, Weight bits=16 bit, Memory=15.45 GB, Evaluation protocol=zero-shot2026.0761.96
GPTQModel=Qwen2.5-VL-32B-Instruct, Weight bits=3 bit, Memory=12.71 GB, Evaluation protocol=zero-shot2026.0761.67
ARB-LLMModel=Qwen2.5-VL-72B-Instruct, Weight bits=1.19 bit, Memory=23.23 GB, Evaluation protocol=zero-shot2026.0761.15
GPTQModel=Qwen2.5-VL-7B-Instruct, Weight bits=3 bit, Memory=2.66 GB, Evaluation protocol=zero-shot2026.0759.93
GPT-4VParams=~1.8T, Training Scale=~10T tokens, Zero-shot evaluation=true2026.0159.9
SAB-LVLMModel=Qwen2.5-VL-32B-Instruct, Weight bits=1.07 bit, Memory=10.28 GB, Evaluation protocol=zero-shot2026.0759.15
BiLLMModel=Qwen2.5-VL-72B-Instruct, Weight bits=1.08 bit, Memory=24.23 GB, Evaluation protocol=zero-shot2026.0758.41
ARB-LLMModel=Qwen2.5-VL-32B-Instruct, Weight bits=1.16 bit, Memory=10.28 GB, Evaluation protocol=zero-shot2026.0757.7
PB-LLMModel=Qwen2.5-VL-72B-Instruct, Weight bits=1.70 bit, Memory=22.99 GB, Evaluation protocol=zero-shot2026.0754.37
Qwen2-VL-2B (baseline)Params=2B, Training Scale=1.2T tokens2026.0150.1
SAB-LVLMModel=Qwen2.5-VL-7B-Instruct, Weight bits=1.07 bit, Memory=2.14 GB, Evaluation protocol=zero-shot2026.0748.11
PB-LLMModel=Qwen2.5-VL-32B-Instruct, Weight bits=1.70 bit, Memory=10.24 GB, Evaluation protocol=zero-shot2026.0747.89
ARB-LLMModel=Qwen2.5-VL-7B-Instruct, Weight bits=1.07 bit, Memory=2.14 GB, Evaluation protocol=zero-shot2026.0747.14
BiLLMModel=Qwen2.5-VL-32B-Instruct, Weight bits=1.08 bit, Memory=10.73 GB, Evaluation protocol=zero-shot2026.0747.04
Video-LLaVA-7BParams=7B, Training Scale=760K videos2026.0145.3
BiLLMModel=Qwen2.5-VL-7B-Instruct, Weight bits=1.08 bit, Memory=2.24 GB, Evaluation protocol=zero-shot2026.0726.56
PB-LLMModel=Qwen2.5-VL-7B-Instruct, Weight bits=1.17 bit, Memory=2.15 GB, Evaluation protocol=zero-shot2026.079.89