Deepfake Detection on UniversalFakeDetect 1.0 (test)

100Accuracy (ProGAN)CNNSpot

Evaluation results

Method
CNNSpot2026.01	100	—	85.2	70.2	85.7	79	91.7	53.5	66.7	48.7	86.3	86.3	60.1	54	55	—	60.8	63.8	65.7	55.6	69.6	54.1	—	—	—
UniFD2026.01	100	—	98.5	94.5	82	99.5	97	66.6	63	57.5	59.5	72	70	94.2	73.8	—	79.1	79.9	78.1	86.8	81.4	94.4	—	—	—
MCAN2026.01	100	—	99.6	98.8	97	99.3	100	94	86.7	68.9	87.3	87.3	70.9	98.8	94.2	—	97.4	97.2	97.1	98.8	93.3	98.5	—	—	—
CNN-Spot2026.01	100	—	93.5	84.5	99.5	89.5	98.2	89	73.8	59.5	98.2	98.4	—	—	—	70.5	80.7	84.9	82.1	70.6	83.6	—	70.6	71	—
F3Net2026.01	100	—	84.3	69.9	99.7	56.7	100	78.8	52.9	46.7	63.4	64.4	—	—	—	74.6	89.8	91	90.9	71.8	76.9	—	73.8	81.7	—
LGrad2026.01	100	—	94	90.7	99.9	79.4	100	67.9	59.4	51.4	63.5	69.6	—	—	—	99.2	93.2	95.1	94.9	97.2	86.4	—	99	99.2	—
UniFD2026.01	100	—	98.1	94.5	86.7	99.3	99.5	91.7	78.5	67.5	83.1	91.1	—	—	—	95.9	93.9	95.1	94.6	88.5	90.1	—	95.8	79.8	—
CLIP-ViT*2026.01	100	—	95.2	95.9	99.5	95	100	93.8	85.2	77.4	93.7	97.5	—	—	—	95.3	79	85.1	82.1	92.3	91.7	—	93.9	90.6	—
NPR2026.01	100	—	99.5	94.5	99.9	88.8	100	84.4	98	100	50.2	50.2	—	—	—	99.9	99.9	99.9	99.9	99.3	92.8	—	99.9	99.9	—
FatFormer2026.01	100	—	100	100	99.8	100	100	98	97.9	81.2	99.8	99.9	—	—	—	99.9	99.1	99.4	99.2	99.8	98.2	—	99.8	99.1	—
RINE2026.01	100	—	100	99.9	99.4	100	100	97.9	97.2	94.9	97.3	99.7	—	—	—	99.9	98.8	99.3	98.9	99.3	98.8	—	99.8	98.3	—
C2P-CLIP2026.01	100	—	100	100	99.5	100	100	98.6	98.9	84.6	99.9	100	—	—	—	100	99.7	99.8	99.8	99.9	98.7	—	100	99.8	—
MPFT2026.01	100	—	100	99.9	99.3	100	100	96.8	100	95.5	100	99.9	—	—	—	99.9	98.1	98.8	98.9	99.6	99	—	99.8	99.4	—
Ojha et al.2025.12	100	—	99.46	99.59	97.24	99.98	99.6	82.45	61.32	79.02	96.72	99	—	—	—	—	99.17	94.74	95.34	94.57	—	92.15	—	—	93.38
Tan et al.2025.12	100	—	90.49	84.55	99.61	77.76	100	90.15	91.21	87.34	67.59	67.72	—	—	—	—	98.56	88.65	92.66	91.22	—	96.94	—	—	90.06
Wang et al.2025.12	100	—	96.83	88.24	98.51	98.09	95.45	66.27	92.76	63.87	98.94	99.52	—	—	—	—	65.99	72.03	76.52	73.22	—	66.74	—	—	81.76
Wang et al.2025.12	100	—	88.01	83.34	99.58	84.38	97	82.23	98.04	65.82	84.69	85.74	—	—	—	—	64.86	79.61	85.4	82.23	—	69.01	—	—	81.93
Wang et al.2025.12	100	—	90.39	83.17	97.69	93.16	96.82	76.79	91.52	74.54	95.2	96.04	—	—	—	—	80.8	81.64	85.18	81.46	—	74.86	—	—	85.11
Wang et al.2025.12	100	—	91.35	84.75	99.7	88.15	98.94	95.95	92.7	75.43	91.69	97.45	—	—	—	—	89.04	91.04	93.76	91.76	—	87.22	—	—	89.93
Frequency-Domain Masking (Wang et al. backbone)2025.12	100	—	92.97	90.06	98.15	97.8	87.94	73.09	89.92	74.57	96.97	98	—	—	—	—	78.21	78.53	82.61	79.27	—	72.44	—	—	85.13
Frequency-Domain Masking (Wang et al. backbone)2025.12	100	—	92.4	92.62	98.33	91.4	99.13	91.35	95.3	77.66	96.54	95.62	—	—	—	—	80.11	83.86	86.95	85.88	—	79.77	—	—	88.84
Frequency-Domain Masking (Wang et al. backbone)2025.12	100	—	93.76	92.19	98.93	97.18	94.93	79.06	95.75	78.67	98.63	98.19	—	—	—	—	81.12	82.81	87.18	83.98	—	77.66	—	—	88.1
Frequency-Domain Masking (Wang et al. backbone)2025.12	100	—	92.1	86.09	96.41	83.76	99.99	93.85	90.81	84	88.84	89.52	—	—	—	—	93.32	90.79	93.67	92.15	—	92.91	—	—	91.04
Gragnaniello et al.2025.12	100	—	83.25	94.3	99.95	91.7	99.99	91.24	92.77	73.69	98.19	97.85	—	—	—	—	87.95	88.24	92.5	89.33	—	86.86	—	—	90.76
Frequency-Domain Masking (Gragnaniello et al. backbone)2025.12	100	—	94.07	97.61	99.92	98.54	99.99	94	95.46	81.47	96.72	95.47	—	—	—	—	94.77	90.32	93.46	91.05	—	94.24	—	—	94.1
FatFormer2026.01	99.9	—	99.3	99.5	97.2	99.4	99.8	93.2	81.1	68	69.5	69.5	76	98.6	94.9	98.7	94.4	94.7	94.2	98.8	90.9	—	—	—	—
FatFormer2026.01	99.9	—	99.3	99.5	97.2	99.4	99.8	93.2	81.1	68	69.5	69.5	76	98.6	94.9	—	94.4	94.7	94.2	98.8	90.9	98.7	—	—	—
FreqNet2026.01	99.9	—	99.6	96.1	99.9	99.7	98.6	99.9	94.4	74.6	80.1	75.7	—	—	—	62.3	99.8	99.8	96.4	77.8	92	—	96.1	100	—
NPR2026.01	99.8	—	95	87.6	96.2	86.6	99.8	76.9	66.9	98.6	50	50	84.6	97.7	98	98.2	96.3	97.2	97.4	87.2	87.6	—	—	—	—
LGrad2026.01	99.8	—	85.4	82.9	94.8	72.5	99.6	58	62.5	50	50.7	50.8	77.5	94.2	95.9	—	87.4	90.7	89.6	88.4	80.3	94.8	—	—	—
NPR2026.01	99.8	—	95	87.6	96.2	86.6	99.8	76.9	66.9	98.6	50	50	84.6	97.7	98	—	96.3	97.2	97.4	87.2	87.6	98.2	—	—	—
Natraj et al.2025.12	99.74	—	80.95	50.61	98.63	53.11	67.99	59.14	68.98	60.42	73.06	87.21	—	—	—	—	92.39	89.32	88.35	82.79	—	89.02	—	—	78.11
DRCT2026.01	99.6	—	93.6	87.6	99.2	90.1	99.9	72.3	67.8	60.5	68.2	59.3	92.9	99.8	99.6	99.8	99.8	99.8	99.9	91.2	88.5	—	—	—	—
DEUA2026.01	99.5	—	94.2	85.3	98.4	90.5	99.5	80.6	72.5	76.4	71.3	74.5	94.8	99.5	99.6	99.9	99.6	99.8	99.8	96.4	91.2	—	—	—	—
F3Net2026.01	99.4	—	76.4	65.3	92.6	58.1	100	63.5	54.2	47.3	51.5	51.5	69.2	68.2	75.4	—	81.7	83.3	83.1	66.3	71.3	68.8	—	—	—
Chai et al.2025.12	98.86	—	72.04	68.79	92.96	55.9	92.06	60.18	65.82	52.87	68.74	67.59	—	—	—	—	88.1	74.54	76.28	75.84	—	84.94	—	—	75.28
FreqNet2026.01	97.9	—	95.8	90.5	97.6	90.2	93.4	97.4	88.9	59	71.9	67.4	86.7	84.6	99.6	—	85.7	97.4	88.2	59.1	85.1	65.6	—	—	—
Patchfor2026.01	80.9	—	72.8	71.7	85.8	66	69.3	76.6	76.2	76.3	74.5	68.5	—	—	—	86.4	85.4	83.7	78.4	75.6	77.7	—	87.1	86.7	—
Chai et al.2025.12	80.88	—	72.84	71.66	85.75	65.99	69.25	76.55	76.19	76.34	74.52	68.52	—	—	—	—	86.4	85.37	83.73	78.38	—	86.72	—	—	77.73
Patchfor2026.01	75	—	69	68.5	79.2	64.2	63.9	75.5	75.1	75.3	72.3	55.3	67.4	76.5	76.1	—	74.8	73.3	68.5	67.9	71.2	75.8	—	—	—
NPR2026.01	57.2	—	73.8	65.2	66	53.5	99	52.9	53	68.4	48.8	50.8	56.2	92.6	92.9	92.7	90.8	86.4	89.9	69.5	71.6	—	—	—	—
Zhang et al.2025.12	55.39	—	100	75.08	55.11	66.08	100	45.18	47.46	57.12	53.61	50.98	—	—	—	—	76.47	68.58	64.58	61.92	—	77.25	—	—	66.21