List your company

Audiovisual Video Captioning on SALMONN 2 (test)

10Miss Ratevideo-SALMONN2-7B
8.12820.76433.446.036Oct 14, 2025Nov 3, 2025Nov 23, 2025Dec 14, 2025Jan 3, 2026Jan 23, 2026Feb 13, 2026

Evaluation results

MethodLinks
video-SALMONN2-7BModality=A + V2025.101012.9—
GPT-4oModality=V2025.101714.2—
Omni-Captioner-7BModality=A + V2025.1017.810.9—
Gemini-2.5-ProSize=-2026.0118.113.331.3
Gemini-2.5-ProModality=A + V2026.0218.113.331.3
Gemini-3-ProModality=A + V2026.0219.116.635.7
Gemini-2.5-FlashSize=-2026.0119.313.933.3
Gemini-2.5-FlashModality=A + V2026.0219.313.933.3
ASID-CaptionerSize=7B, Modality=A + V2026.0220.515.435.9
AVoCaDOSize=7B2026.0121.116.237.3
DiaDemSize=7B2026.0121.115.536.5
AVoCaDOSize=7B, Modality=A + V2026.0221.116.237.3
video-SALMONN-2Size=7B2026.0121.217.638.8
video-SALMONN-2Size=7B, Modality=A + V2026.0221.217.638.8
Gemini-3-ProSize=-2026.0121.414.235.6
Gemini 1.5 ProModality=A + V2025.1021.816.5—
Qwen2.5-VL-7BModality=V2025.1021.917.4—
LLaVA-OneVision-7BModality=V2025.1023.327.4—
ASID-CaptionerSize=3B, Modality=A + V2026.0223.418.341.7
Qwen2.5-Omni-7BModality=A + V2025.1026.321.7—
InternVideo2.5-7BModality=V2025.1030.815—
Qwen3-Omni-CaptionerSize=30B-A3B2026.013116.647.6
Qwen3-Omni-CaptionerSize=30B-A3B, Modality=A + V2026.023116.647.6
UGC-VideoCaptionerSize=3B2026.0131.61748.6
UGC-VideoCaptionerSize=3B, Modality=A + V2026.0231.61748.6
Qwen3-Omni-InstructSize=30B-A3B2026.013213.645.6
Qwen3-Omni-InstructSize=30B-A3B, Modality=A + V2026.023213.645.6
Qwen2.5-VLSize=7B, Modality=V2026.0240.51757.5
Qwen2.5-OmniSize=7B2026.0141.715.457.1
Qwen2.5-OmniSize=7B, Modality=A + V2026.0241.715.457.1
MiniCPM-o-2.6Size=8B, Modality=A + V2026.0242.214.356.5
VideoLLaMA3-7BModality=V2025.1044.911.6—
ARC-Hunyuan-VideoSize=7B, Modality=A + V2026.0245.712.558.2
HumanOmniV2Size=7B, Modality=A + V2026.0249.212.361.6
video-SALMONN-13BModality=A + V2025.1052.126.6—
InternVL3.5Size=8B, Modality=V2026.0253.825.579.4
VideoLLaMA2-7BModality=A + V2025.1056.88.9—