List your company

Video Question Answering on RAVEN-QA Simple Queries

100AccuracyRAVEN
9.5233.0156.579.99Jun 23, 2026

Evaluation results

MethodLinks
RAVENEmbedder=Seed1.6-Embed, VLM=Gp32026.06100
VLM OnlyVLM=Gp32026.06100
RAVENEmbedder=Seed1.6-Embed, VLM=Gf2.52026.0698.1
RAVENEmbedder=Seed1.6-Embed, VLM=GPT-52026.0698.1
RAVENEmbedder=Seed1.6-Embed, VLM=GPT-4o2026.0698.1
RAVENEmbedder=QQMM-v2, VLM=Gp32026.0698.1
RAVENEmbedder=QQMM-v2, VLM=Gp2.52026.0698.1
RAVENEmbedder=SigLIP, VLM=Gp32026.0698.1
RAVENEmbedder=SigLIP, VLM=Gp2.52026.0698.1
RAVENEmbedder=CLIP-B, VLM=GPT-52026.0698.1
VLM OnlyVLM=Gp2.52026.0698.1
RAVENEmbedder=Seed1.6-Embed, VLM=Gp2.52026.0696.3
RAVENEmbedder=QQMM-v2, VLM=Gf2.52026.0696.3
RAVENEmbedder=QQMM-v2, VLM=GPT-52026.0696.3
RAVENEmbedder=QQMM-v2, VLM=GPT-4o2026.0696.3
RAVENEmbedder=CLIP-B, VLM=Gp2.52026.0696.3
VLM OnlyVLM=Gf2.52026.0696.3
Caption (4o-mini)VLM=Gp2.52026.0696.3
RAVENEmbedder=SigLIP, VLM=Gf2.52026.0694.4
RAVENEmbedder=SigLIP, VLM=GPT-52026.0694.4
RAVENEmbedder=SigLIP, VLM=GPT-4o2026.0694.4
RAVENEmbedder=CLIP-B, VLM=Gp32026.0694.4
RAVENEmbedder=CLIP-B, VLM=Gf2.52026.0694.4
VLM OnlyVLM=GPT-4o2026.0694.4
RAVENEmbedder=QQMM-v2, VLM=Q2.5-32b2026.0692.6
RAVENEmbedder=QQMM-v2, VLM=Q3-32b2026.0692.6
Caption (4o-mini)VLM=Gf2.52026.0692.6
RAVENEmbedder=QQMM-v2, VLM=Embedder Only2026.0690.7
Caption (4o-mini)VLM=GPT-52026.0690.7
Caption (4o-mini)VLM=Q3-32b2026.0690.7
RAVENEmbedder=CLIP-B, VLM=GPT-4o2026.0688.9
Caption (4o-mini)VLM=GPT-4o2026.0688.9
RAVENEmbedder=Seed1.6-Embed, VLM=Gm3-12b2026.0687
Caption (4o-mini)VLM=Gp32026.0687
RAVENEmbedder=Seed1.6-Embed, VLM=Embedder Only2026.0685.2
RAVENEmbedder=QQMM-v2, VLM=Gm3-27b2026.0685.2
RAVENEmbedder=Seed1.6-Embed, VLM=Q2.5-32b2026.0683.3
RAVENEmbedder=Seed1.6-Embed, VLM=Q3-32b2026.0683.3
RAVENEmbedder=SigLIP, VLM=Embedder Only2026.0683.3
VLM OnlyVLM=GPT-52026.0683.3
RAVENEmbedder=QQMM-v2, VLM=Gm3-12b2026.0681.5
RAVENEmbedder=SigLIP, VLM=Q2.5-32b2026.0679.6
RAVENEmbedder=CLIP-B, VLM=Gm3-27b2026.0677.8
RAVENEmbedder=SigLIP, VLM=Gm3-27b2026.0675.9
RAVENEmbedder=SigLIP, VLM=Gm3-12b2026.0675.9
RAVENEmbedder=Seed1.6-Embed, VLM=Gm3-27b2026.0674.1
RAVENEmbedder=CLIP-B, VLM=Gm3-12b2026.0674.1
RAVENEmbedder=CLIP-B, VLM=Embedder Only2026.0674.1
RAVENEmbedder=SigLIP, VLM=Q3-32b2026.0672.2
RAVENEmbedder=CLIP-B, VLM=Q3-32b2026.0672.2
Caption (4o-mini)VLM=Q2.5-32b2026.0668.5
RAVENEmbedder=CLIP-B, VLM=Q2.5-32b2026.0659.3
Caption (4o-mini)VLM=Q2.5-7b2026.0659.3
RAVENEmbedder=Seed1.6-Embed, VLM=Q2.5-7b2026.0631.5
RAVENEmbedder=SigLIP, VLM=Q2.5-7b2026.0631.5
VLM OnlyVLM=Q2.5-32b2026.0629.6
Caption (4o-mini)VLM=Gm3-12b2026.0627.8
VLM OnlyVLM=Q3-32b2026.0625.9
RAVENEmbedder=QQMM-v2, VLM=Q2.5-7b2026.0622.2
RAVENEmbedder=CLIP-B, VLM=Q2.5-7b2026.0618.5
Caption (4o-mini)VLM=Gm3-27b2026.0618.5
VLM OnlyVLM=Gm3-12b2026.0614.8
VLM OnlyVLM=Gm3-27b2026.0613
VLM OnlyVLM=Q2.5-7b2026.0613