Instruction-guided Image Editing on KRIS-Bench

85.61Attribute AccuracyGemini-3-Pro-Image

Evaluation results

Method
Gemini-3-Pro-Image2026.04	85.61	87.17	88.06	86.36	85.25	83.64	84.03	81.42	93.22	86.48	85.31
GPT-4o2026.04	83.17	79.08	68.25	79.8	85.5	80.06	81.37	71.56	85.08	78.32	80.09
DDA-Thinker2026.04	82.16	84.58	75	81.52	84.71	77.77	79.44	74.45	84.67	78.88	79.94
ThinkRL-Edit2026.04	81.02	81.45	—	81.13	75.67	71.25	72.31	49.07	79.71	57.44	71.65
DDA-Thinker2026.04	79.14	79.33	74.43	78.41	79.86	76.93	77.64	68.97	80.11	73.83	76.99
Gemini-2.5-Flash-Image2026.04	78.97	80.17	86.49	80.48	83.7	76.34	78.11	63.96	90.44	75.31	78.17
EditThinker2026.04	78.48	73.83	—	77.24	76.2	70.69	72.02	65.23	66.89	65.94	71.91
Qwen-Image-Edit-25112026.04	74.85	84.5	68.13	75.89	66.1	59.54	61.15	58.42	79.61	67.55	67.14
DDA-Thinker2026.04	72.79	77.92	73.09	73.98	76.9	70.6	72.13	55.43	89.39	70.07	72.21
Uni-CoT2026.04	72.76	72.87	67.1	71.85	70.81	66	67.16	53.43	73.93	63.68	68
Qwen-Image-Edit-25092026.04	72.44	79.1	55.07	71.05	66.14	57.75	59.78	49.83	77.61	61.94	63.71
DDA-Thinker2026.04	71.55	77	72.52	72.92	76.55	68.15	70.18	54.47	86.56	68.25	70.55
BAGEL-Think2026.04	67.42	68.33	58.67	66.18	63.55	61.4	61.92	48.12	50.22	49.02	60.18
Gemini 2.02026.04	66.33	63.33	63.92	65.26	68.19	56.94	59.65	54.13	71.67	62.9	62.41
FLUX.1-Kontext-Dev2026.04	64.83	60.92	—	53.28	48.94	50.81	50.36	46.06	39	42.53	49.54
BAGEL2026.04	64.27	62.42	42.45	60.26	55.4	56.01	55.86	52.54	50.56	51.69	56.21
Step1X-Edit v1.12026.04	64.17	61.75	—	53.05	52.06	55.06	54.34	52.56	36.75	44.66	51.59
ByteMorph2026.04	61.17	62	—	51.27	45.5	47.38	46.92	32	31.33	31.67	44.85
OmniGen22026.04	59.92	52.25	54.75	57.36	47.56	43.12	44.2	32.5	63.08	47.79	49.71
UniWorld-V12026.04	58.17	54.5	63	47.71	47.5	43.94	44.8	42	53.83	47.92	50.27
Step1X-Edit2026.04	55.5	51.75	—	45.52	44.69	49.06	48.01	40.88	22.75	31.82	43.29
MagicBrush2026.04	53.92	39.58	—	41.84	42.94	38.06	39.24	30	23.08	26.54	37.15
HiDream-E12026.04	52.75	49.42	—	43.31	52.56	49.25	50.05	45.19	30.08	37.64	44.72
Emu22026.04	51.5	48.83	22.17	45.4	34.69	38.44	37.54	24.81	45	34.91	39.7
AnyEdit2026.04	47.67	45.17	—	39.26	38.56	42.94	41.88	36.56	26.92	31.74	38.55
OmniGen2026.04	37.92	28.25	21.83	33.11	30.63	27.19	28.02	11.94	35.83	23.89	28.85
InstructPix2Pix2026.04	30.33	21.33	—	23.33	22.56	26.56	25.59	19.81	14.75	17.28	22.82