Hugging Face's logo

Datasets:
dispatchAI
/
speed-benchmark

Modalities:
Tabular
Text
Formats:
json
Size:
< 1K
Tags:
speed
benchmark
mobile
dispatchai
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
3morixd commited on
Commit
d777943
·
verified ·
1 Parent(s): 5339a5c

Upload speeds.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. speeds.json +156 -0
speeds.json ADDED
@@ -0,0 +1,156 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "model": "SmolLM2-135M-Instruct-mobile",
4
+ "cpu_tps": 59.7,
5
+ "phone_tps": 46.0,
6
+ "size_mb": 0,
7
+ "chat_format": "llama-3"
8
+ },
9
+ {
10
+ "model": "TinyLlama-1.1B-Chat-mobile-int4",
11
+ "cpu_tps": 17.6,
12
+ "phone_tps": "",
13
+ "size_mb": 637.8,
14
+ "chat_format": "chatml"
15
+ },
16
+ {
17
+ "model": "Qwen2.5-0.5B-Instruct-Server-mobile",
18
+ "cpu_tps": 16.7,
19
+ "phone_tps": "",
20
+ "size_mb": 482.3,
21
+ "chat_format": "chatml"
22
+ },
23
+ {
24
+ "model": "Qwen2.5-0.5B-Chinese-mobile",
25
+ "cpu_tps": 16.4,
26
+ "phone_tps": "",
27
+ "size_mb": 400.6,
28
+ "chat_format": "chatml"
29
+ },
30
+ {
31
+ "model": "Qwen2.5-Math-1.5B-mobile",
32
+ "cpu_tps": 15.7,
33
+ "phone_tps": "",
34
+ "size_mb": 940.4,
35
+ "chat_format": "chatml"
36
+ },
37
+ {
38
+ "model": "Qwen2.5-0.5B-Instruct-mobile-int4",
39
+ "cpu_tps": 12.5,
40
+ "phone_tps": 23.2,
41
+ "size_mb": 0,
42
+ "chat_format": "chatml"
43
+ },
44
+ {
45
+ "model": "Llama-3.2-1B-Instruct-Q4-mobile",
46
+ "cpu_tps": 11.3,
47
+ "phone_tps": 5.4,
48
+ "size_mb": 0,
49
+ "chat_format": "chatml"
50
+ },
51
+ {
52
+ "model": "TinyLlama-1.1B-Chat-Q5-mobile",
53
+ "cpu_tps": 10.9,
54
+ "phone_tps": "",
55
+ "size_mb": 746.7,
56
+ "chat_format": "chatml"
57
+ },
58
+ {
59
+ "model": "Qwen2.5-1.5B-Instruct-Q8-mobile",
60
+ "cpu_tps": 8.7,
61
+ "phone_tps": "",
62
+ "size_mb": 1570.3,
63
+ "chat_format": "chatml"
64
+ },
65
+ {
66
+ "model": "Phi-3.5-mini-Instruct-mobile",
67
+ "cpu_tps": 8.6,
68
+ "phone_tps": "",
69
+ "size_mb": 2282.4,
70
+ "chat_format": "chatml"
71
+ },
72
+ {
73
+ "model": "Gemma-2-2B-IT-mobile",
74
+ "cpu_tps": 8.2,
75
+ "phone_tps": "",
76
+ "size_mb": 1629.4,
77
+ "chat_format": "gemma"
78
+ },
79
+ {
80
+ "model": "Qwen2.5-0.5B-Coder-mobile",
81
+ "cpu_tps": 8.0,
82
+ "phone_tps": "",
83
+ "size_mb": 379.4,
84
+ "chat_format": "chatml"
85
+ },
86
+ {
87
+ "model": "Qwen2.5-Coder-1.5B-mobile",
88
+ "cpu_tps": 7.8,
89
+ "phone_tps": "",
90
+ "size_mb": 940.4,
91
+ "chat_format": "chatml"
92
+ },
93
+ {
94
+ "model": "Qwen2.5-1.5B-Instruct-mobile-int4",
95
+ "cpu_tps": 7.7,
96
+ "phone_tps": "",
97
+ "size_mb": 1065.6,
98
+ "chat_format": "chatml"
99
+ },
100
+ {
101
+ "model": "Phi-3.5-mini-instruct-Q5-mobile",
102
+ "cpu_tps": 7.0,
103
+ "phone_tps": "",
104
+ "size_mb": 2684.9,
105
+ "chat_format": "chatml"
106
+ },
107
+ {
108
+ "model": "Gemma-2-2B-IT-Q5-mobile",
109
+ "cpu_tps": 5.7,
110
+ "phone_tps": "",
111
+ "size_mb": 1834.2,
112
+ "chat_format": "gemma"
113
+ },
114
+ {
115
+ "model": "Llama-3.2-1B-Instruct-Q6-mobile",
116
+ "cpu_tps": 5.0,
117
+ "phone_tps": "",
118
+ "size_mb": 974.5,
119
+ "chat_format": "llama-3"
120
+ },
121
+ {
122
+ "model": "Llama-3.2-1B-Instruct-mobile",
123
+ "cpu_tps": 4.9,
124
+ "phone_tps": "",
125
+ "size_mb": 1259.9,
126
+ "chat_format": "llama-3"
127
+ },
128
+ {
129
+ "model": "Llama-3.2-3B-Instruct-mobile",
130
+ "cpu_tps": 4.8,
131
+ "phone_tps": "",
132
+ "size_mb": 1925.8,
133
+ "chat_format": "llama-3"
134
+ },
135
+ {
136
+ "model": "Qwen2.5-1.5B-Instruct-Q5-mobile",
137
+ "cpu_tps": 4.4,
138
+ "phone_tps": "",
139
+ "size_mb": 1072.9,
140
+ "chat_format": "chatml"
141
+ },
142
+ {
143
+ "model": "Qwen2.5-Coder-7B-mobile",
144
+ "cpu_tps": 3.0,
145
+ "phone_tps": "",
146
+ "size_mb": 4466.1,
147
+ "chat_format": "chatml"
148
+ },
149
+ {
150
+ "model": "Llama-3.2-3B-Instruct-Q5-mobile",
151
+ "cpu_tps": 2.6,
152
+ "phone_tps": "",
153
+ "size_mb": 2214.6,
154
+ "chat_format": "llama-3"
155
+ }
156
+ ]