README.md

metadata

license: mit

ggml files of thenlper/gte-large

You can use this ggml for https://github.com/skeskinen/bert.cpp

gte-large

Data Type	STSBenchmark	eval time	EmotionClassification	eval time
f32	0.8606	127.58	0.5060	199.61
f16	0.8606	103.89	0.5060	169.68
q4_0	0.8589	80.85	0.5037	157.05
q4_1	0.8605	90.13	0.5107	162.59

all-MiniLM-L12-v2

Data Type	STSBenchmark	eval time	EmotionClassification	eval time
f32	0.8306	13.36	0.4117	21.23
f16	0.8306	11.51	0.4119	20.08
q4_0	0.8310	11.27	0.4183	20.81
q4_1	0.8325	12.37	0.4093	19.38

all-MiniLM-L6-v2

Data Type	STSBenchmark	eval time	EmotionClassification	eval time
f32	0.8201	6.83	0.4082	11.34
f16	0.8201	6.17	0.4085	10.28
q4_0	0.8175	5.45	0.3911	10.63
q4_1	0.8223	6.79	0.4027	11.41

bert-base-uncased

Data Type	STSBenchmark	eval time	EmotionClassification	eval time
f32	0.4738	52.38	0.3361	88.56
f16	0.4739	33.24	0.3361	55.86
q4_0	0.4940	33.93	0.3375	57.82
q4_1	0.4612	36.86	0.3318	59.63