List your company

Video Understanding on Video-Holmes (Accuracy)

58.3AccuracyAdaTooler-V-7B
12.74824.57436.448.226Dec 18, 2025Jan 6, 2026Jan 25, 2026Feb 14, 2026Mar 5, 2026Mar 24, 2026Apr 13, 2026

Evaluation results

MethodLinks
AdaTooler-V-7BFrames=1282025.1258.3
AdaTooler-V-7BFrames=642025.1256.4
AdaTooler-V-7BFrames=322025.1255.6
Gemini 1.5 proModel Category=Proprietary Models, Venue=-2026.0345.7
Video-Thinker2025.1243.2
GPT-4o2025.1242
GPT-4oModel Category=Proprietary Models, Venue=-2026.0342
VST-7BModel Category=Open-source Online Models, Venue=-2026.0341.9
Gemini 1.5 Pro2025.1241.2
InternVL3-8B (MERIT)Model=InternVL3-8B, Variant=MERIT2026.0437.2
InternVL3-8B (RANDOM-K)Model=InternVL3-8B, Variant=RANDOM-K2026.0436.7
Video-R12025.1236.5
Video-R1-7BModel Category=Open-source Offline Models, Venue=NeurIPS’252026.0336.5
InternVL3-8B (BASE)Model=InternVL3-8B, Variant=BASE2026.0436.5
InternVL3-8B (ALL-LAYER)Model=InternVL3-8B, Variant=ALL-LAYER2026.0436.2
Qwen2.5-VL-7B-InstructFrames=1282025.1233.5
VideoChat-R12025.1233
Qwen3-VL-4B (MERIT)Model=Qwen3-VL-4B, Variant=MERIT2026.0432.9
Qwen3-VL-4B (ALL-LAYER)Model=Qwen3-VL-4B, Variant=ALL-LAYER2026.0431.9
Qwen3-VL-4B (BASE)Model=Qwen3-VL-4B, Variant=BASE2026.0431.5
Qwen3-VL-4B (RANDOM-K)Model=Qwen3-VL-4B, Variant=RANDOM-K2026.0431
Qwen2.5-VL-7B-InstructFrames=642025.1229.9
Qwen2.5-VL-7B-InstructFrames=322025.1227.8
LongVA-7B (MERIT)Model=LongVA-7B, Variant=MERIT2026.0418.9
LongVA-7B (ALL-LAYER)Model=LongVA-7B, Variant=ALL-LAYER2026.0415.4
LongVA-7B (BASE)Model=LongVA-7B, Variant=BASE2026.0415.2
LongVA-7B (RANDOM-K)Model=LongVA-7B, Variant=RANDOM-K2026.0414.5