Knowledge-grounded reasoning on WISE

89Overall ScoreGPT-4o

Evaluation Results

Method
GPT-4o2025.12	89	94	64	98	93	98	95
LVRPO2026.03	85	86	78	93	89	87	79
GPT4o2026.01	80	81	71	89	83	79	74
GPT4o2025.12	80	81	71	89	83	79	74
GPT-4o2026.03	80	81	71	89	83	79	74
Seedream 4.02025.12	78	78	73	85	79	84	67
GenAgent2026.01	72	78	67	78	72	71	55
Bagel2026.01	70	76	69	75	65	75	58
BAGEL2025.12	70	76	69	75	65	75	58
BAGEL2026.03	70	76	69	75	65	75	58
GenAgent2026.01	69	75	69	72	61	65	50
GenAgent2026.01	67	75	66	72	55	60	49
STAR-7B2025.12	66	61	67	61	74	69	66
LongCat-Image2025.12	65	66	61	72	66	72	49
StruVis2026.03	65	75	59	63	58	58	51
StruVis2026.03	65	74	59	62	59	56	51
GenAgent2026.01	64	68	61	68	60	65	47
GenAgent2026.01	63	70	63	68	53	61	41
Qwen-Image2026.01	62	62	63	77	57	75	40
BLIP3-o2025.12	62	—	—	—	—	—	—
Qwen-Image2025.12	62	62	63	77	57	75	40
Text-Only2026.03	58	66	51	62	52	51	47
HunyuanImage-3.02025.12	57	58	57	70	56	63	31
Interleaved2026.03	57	62	55	61	53	51	47
PromptEnhancer2026.01	56	54	60	69	46	62	39
W/O.2026.03	56	62	49	61	50	54	41
Text-Only2026.03	56	62	49	61	50	56	44
FLUX.1-dev2026.01	55	56	57	67	45	54	41
MetaQuery-XL2025.12	55	56	55	62	49	63	41
MetaQuery-XL2025.12	55	56	55	62	49	63	41
Interleaved2026.03	55	61	50	62	50	52	42
MetaQuery-XL2026.03	55	56	55	62	49	63	41
T2I-R12026.01	54	56	55	63	54	55	30
W/O.2026.03	53	61	45	61	44	52	41
Bagel2026.01	52	44	55	68	44	60	39
STAR-3B2025.12	52	58	54	48	49	51	54
FLUX.1-dev2025.12	50	48	58	62	42	51	35
FLUX.1-dev2025.12	50	48	58	62	42	51	35
FLUX.1-dev2026.03	50	48	58	62	42	51	35
Playground-v2.52025.12	49	49	58	55	43	48	33
playground-v2.52026.03	49	49	58	55	43	48	33
PixArt-α2025.12	47	45	50	48	49	56	34
PixArt-Alpha2026.03	47	45	50	48	49	56	34
SD-3.5-large2025.12	46	44	50	58	44	52	31
SD-3.5-large2025.12	46	44	50	58	44	52	31
SD3.5-large2026.03	46	44	50	58	44	52	31
SD-3.5-medium2025.12	45	43	50	52	41	53	33
SD-XL2025.12	43	43	48	47	44	45	27
SDXL2026.03	43	43	48	47	44	45	27
SD-3-medium2025.12	42	42	44	48	39	47	29
FLUX.1-schnell2025.12	40	39	44	50	31	44	26
Emu32025.12	39	34	45	48	41	45	27
Liquid2025.12	39	34	45	48	41	45	27
Emu32025.12	39	34	45	48	41	45	27
Emu32026.03	39	34	45	48	41	45	27
Janus Pro 7B2026.01	35	30	37	49	36	42	26
Janus-Pro-7B2025.12	35	30	37	49	36	42	26
Janus-Pro-7B2025.12	35	30	37	49	36	42	26
Show-o-5122025.12	35	28	40	48	30	46	30
Show-o-5122026.03	35	28	40	48	30	46	30
Janus-Pro-7B2026.03	35	30	37	49	36	42	26
SDv1.52026.03	32	34	35	32	28	29	21
VILA-U2026.03	31	26	33	37	35	39	23
Orthus-7B-instruct2025.12	27	23	31	38	28	31	20
Janus-Pro-1B2025.12	26	20	28	45	24	32	16
Janus-1.3B2025.12	23	16	26	35	28	30	14
Janus2026.03	23	16	26	35	28	30	14
JanusFlow2025.12	18	13	26	28	20	19	11