Open-ended VQA on MMOral-OPG

59.7Teeth AccuracyOPGAgent

Evaluation Results

Method
OPGAgent2026.02	59.7	52.7	71.8	72.1	—	—	62.53
OralAgent2026.04	55.91	56.74	59.14	76.25	70	61	61
Gemini-3-Flash2026.02	54.1	42.6	66.2	59.7	—	—	48.47
OralGPT-Omni2026.02	53.5	44.6	69.7	77.5	—	—	59.06
OralGPT-Plus-7B2026.03	45.34	33.56	45.58	75.38	34.64	32.7	45.35
DentalGPT2026.02	42.3	40.5	47.2	65.1	—	—	48.27
GPT-52026.03	39.84	29.42	44.08	77.38	40.06	28.16	42.34
GPT-52025.11	39.77	29.32	44.05	78.56	40.12	28.2	42.42
GPT-52026.04	39.77	29.32	44.05	78.56	40.12	28.2	42.42
OralGPT-Omni2025.11	37.26	43.94	55.34	70.5	38.57	37.9	45.31
OralGPT-Omni2026.04	37.26	43.94	55.34	70.5	38.57	37.9	45.31
GPT-5.22026.02	37.2	32.4	52.1	65.9	—	—	43.58
MDAgents2026.04	36.19	31.52	54.36	64.94	39.29	26.9	40.5
HuatuoGPT-V2026.03	35.18	24.92	36.32	65.69	25.48	24.8	36.02
OralGPT-Plus-7B-SFT2026.03	35.02	19.32	29.02	62.37	20.12	11.9	31.44
MedRAX2026.04	32.35	18.11	42.64	59.63	45.36	32.1	36.73
Qwen2.5-VL-3B-Instruct2026.03	32.01	20.91	27.55	54.5	30.12	5.7	28.79
GPT-4V2025.11	31.46	23.79	39.51	69.81	34.29	43.7	39.38
GPT-4V2026.04	31.46	23.79	39.51	69.81	34.29	43.7	39.38
MedAgents2026.04	30.94	24.17	44.17	67	37.38	15.4	34.71
HealthGPT-XL322025.11	30.64	25.83	27.98	51.12	17.02	8	27.8
HealthGPT2026.03	30.64	25.83	27.98	51.12	17.02	8	27.8
HealthGPT-XL322026.04	30.64	25.83	27.98	51.12	17.02	8	27.8
MedGemma-4B2026.03	29.69	11.26	28.12	41.88	22.62	39.4	30.45
LLaVA-OneVision2026.03	28.79	13.64	30.12	55.12	20.36	23.2	29.43
Claude-sonnet-4-52026.03	28.29	20.45	48.83	71.56	26.67	37.7	37.68
Gemini-2.5-Flash2026.03	28.21	24.64	31.95	47.74	13.05	16.73	27.89
Gemini-2.5-Flash2025.11	28.04	24.77	31.9	47.81	12.98	16.7	27.84
Gemini-2.5-Flash2026.04	28.04	24.77	31.9	47.81	12.98	16.7	27.84
GLM-4.5v2026.03	27.53	22.2	46.2	59.56	22.14	42.7	35.95
Qwen3-VL-235b-a22b-thinking2026.03	27.42	26.36	39.34	58.5	40.48	37.7	35.78
Doubao-seed-1-62026.03	27.19	18.18	41.72	59.44	22.38	40.2	34.2
MedDr2025.11	22.99	32.58	29.57	52.44	20.95	8.7	26.2
MedDr2026.03	22.99	32.58	29.57	52.44	20.95	8.7	26.2
MedDr2026.04	22.99	32.58	29.57	52.44	20.95	8.7	26.2
Grok-42026.03	22.48	19.08	32.82	54.81	25.36	38.3	30.97
MedVLM-R12025.11	22.42	13.71	24.42	43.88	13.57	25.8	24.7
MedVLM-R12026.03	22.42	13.71	24.42	43.88	13.57	25.8	24.7
MedVLM-R12026.04	22.42	13.71	24.42	43.88	13.57	25.8	24.7
OralGPT-Plus-3B-SFT2026.03	22.17	13.71	16.44	35.25	9.88	19.8	21.02
OralGPT-Plus-3B2026.03	21.58	17.35	21.29	35.69	16.07	31.7	24.48
GLM-4V-9B2025.11	20.94	9.7	18.77	26.62	12.74	21.3	19.74
GLM-4V-9B-Thinking2026.04	20.94	9.7	18.77	26.62	12.74	21.3	19.74
Qwen2.5-VL-7B-Instruct2026.03	17.01	16.1	11.18	29.41	9.07	8.2	15.92
Deepseek-VL-7b-chat2025.11	16.48	7.5	13.44	34.56	9.52	9.6	15.95
Deepseek-VL-7b-chat2026.04	16.48	7.5	13.44	34.56	9.52	9.6	15.95
MMedAgent2026.04	16	10.08	15.15	39.81	8.93	0.5	15.19
Qwen2.5-VL-72B2025.11	13.9	15.83	15.4	27.12	7.38	11.5	15.38
Qwen2.5-VL-72B2026.04	13.9	15.83	15.4	27.12	7.38	11.5	15.38
Mini-o32026.03	8.72	1.97	15.15	46.62	13.93	9.8	14.72
Qwen-Max-VL2025.11	2.1	4.47	7.06	11.62	7.98	5.5	5.29
Qwen-Max-VL2026.04	2.1	4.47	7.06	11.62	7.98	5.5	5.29
LLaVA-Med2026.03	0.92	1.52	0	0	0	24.5	4.76
LLaVA-Med2025.11	0.91	1.52	0	0	0	24.5	4.76
LLaVA-Med2026.04	0.91	1.52	0	0	0	24.5	4.76