Large Vision-Language Modeling on Tiny LVLM (test) benchmark leaderboard · SOTA2 Research

Large Vision-Language Modeling on Tiny LVLM (test)

64.2VRBard

Evaluation Results

Method	Links
Bard2023.12		64.2	319.6	57	68.1	59.6	70.7
Bard2023.09		64.2	319.6	57	68.1	59.6	70.7
Qwen-VL-Chat2023.12		62.4	316.8	54.5	55.1	54.8	90
Qwen-VL-Chat2023.09		62.4	316.8	54.5	55.1	54.8	90
InternVL-Chat2023.12		56.4	327.6	52.3	68	62	89
InternLM-XComposer2023.12		55.8	322.5	53.8	64.1	61.8	87
InternLM-XComposer2023.09		55.8	322.5	53.8	64.1	61.8	87
LLaVA-1.52023.12		55.6	307.2	49	57	57.2	88.3
LLaVA-1.52023.09		55.6	307.2	49	57	57.2	88.3
Lynx2023.12		52.2	279.2	65.8	17.6	57.4	86.3
Lynx2023.09		52.2	279.2	65.8	17.6	57.4	86.3
InstructBLIP2023.12		46.7	300.6	48	61.7	59.2	85
InstructBLIP2023.09		46.7	300.6	48	61.7	59.2	85
BLIP-22023.12		44.9	284.7	49	64.1	44	82.7
BLIP22023.09		44.9	284.7	49	64.1	44	82.7
LLaMA-Adapter-V22023.12		43.5	229.2	46.8	22.3	56	60.7
LLaMA-A-V22023.09		43.5	229.2	46.8	22.3	56	60.7
LLaVA2023.12		41.6	197	38.3	18.7	49.4	49
Otter2023.12		41.6	216.4	37	15.1	52.4	74
LLaVA2023.09		41.6	197	38.3	18.7	49.4	49
Otter2023.09		41.6	216.4	37	15.1	52.4	74
MiniGPT-42023.12		37.6	192.6	37.8	17.6	49	50.7
MiniGPT-42023.09		37.6	192.6	37.8	17.6	49	50.7
VisualGLM2023.12		37.3	211.9	36.3	46.9	37.6	54
VisualGLM2023.09		37.3	211.9	36.3	46.9	37.6	54