Back to Benchmarks

EVA Benchmark
pathology
H&E
radiology

Comprehensive evaluation framework for pathology foundation models covering patch-level classification, slide-level analysis, and segmentation tasks.

15 models evaluated
13 tasks
Organs:
breast
colon
prostate
multi-organ

Detailed Results

Model
Average
rank
Average
metric
BACH
Balanced Accuracy
BreakHis
Balanced Accuracy
Camelyon16
Balanced Accuracy
Camelyon16 (test)
Balanced Accuracy
CoNSeP
Dice
CRC
Balanced Accuracy
Gleason
Balanced Accuracy
MHIST
Balanced Accuracy
MoNuSAC
Dice
PANDA
Balanced Accuracy
PANDA (test)
Balanced Accuracy
PCam
Balanced Accuracy
PCam (test)
Balanced Accuracy
3.150.8150.8830.8210.9740.8610.6400.9670.7830.8610.6690.6890.6460.9330.938
4.000.8120.9150.8590.9730.8490.6300.9650.7750.8240.6420.6780.6570.9440.950
4.620.8070.7590.8010.9290.8270.6440.9550.7700.8430.6850.6890.6710.9320.943
5.620.8020.7850.7850.9570.8330.6280.9440.7500.8430.6590.6910.6590.9360.937
5.690.8030.9040.8190.9180.8490.6230.9660.8000.8040.6590.6890.6490.9290.929
6.920.7930.8100.7350.9240.8230.6460.9320.7640.8390.6690.6610.6340.9390.955
6.920.7930.8660.8320.8660.8140.6460.9590.7440.8290.6800.6840.6400.9120.915
7.540.7930.8640.8370.8980.8160.6350.9380.7490.8300.6700.6740.6450.9140.911
7.850.7950.7590.8270.8810.8150.6260.9510.7240.8290.6800.6700.6530.9350.945
9.230.7840.8350.8110.8970.8190.6060.9570.7300.8370.6390.6680.6320.9050.906
10.920.7750.7290.7170.8950.7980.6270.9400.7290.8030.6350.6740.6440.9160.920
11.230.7690.8020.7080.8720.8040.6370.9520.7180.8230.6530.6760.6010.8940.886
11.310.7740.7320.7130.9220.8050.6270.9390.7570.7770.6440.6640.6250.9180.894
12.000.7670.8120.7340.8780.7880.6030.9400.6980.8310.6350.6670.6210.9050.902
13.000.7610.7830.7420.8460.7710.6020.9400.7500.7810.6290.6680.6100.8940.897