File size: 3,816 Bytes
0c0f086
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b63ff12
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0c0f086
 
9f46468
 
 
b63ff12
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9f46468
 
0c0f086
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37a6a80
 
 
10d93f3
37a6a80
 
 
 
0c0f086
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
"""Module containing performance results for the Llama-Spark model."""

results_llama_spark = {
    "name": "Llama-Spark",
    "modelType": "Llama 3.1 8B",
    "configurations": [
        {
            "instanceType": "g5.2xlarge",
            "quantization": "none",
            "container": "TGI 2.2.0",
            "status": "OK",
            "tokensPerSecond": "29",
            "notes": "4K/8K fails",
        },
        {
            "instanceType": "g5.12xlarge",
            "quantization": "none",
            "container": "TGI 2.2.0",
            "status": "OK",
            "tokensPerSecond": "85",
            "notes": '"MAX_INPUT_TOKENS": "16384", "MAX_TOTAL_TOKENS": "32768",',
        },
        {
            "instanceType": "g5.48xlarge",
            "quantization": "none",
            "container": "TGI 2.2.0",
            "status": "OK",
            "tokensPerSecond": "105",
            "notes": '"MAX_INPUT_TOKENS": "20480", "MAX_TOTAL_TOKENS": "40960"\n\n32K/64K fails',
        },
        {
            "instanceType": "g6.12xlarge",
            "quantization": "none",
            "container": "TGI 2.2.0",
            "status": "OK",
            "tokensPerSecond": "51",
            "notes": "same as g5?",
        },
        {
            "instanceType": "g6.48xlarge",
            "quantization": "none",
            "container": "TGI 2.2.0",
            "status": "OK",
            "tokensPerSecond": "81",
            "notes": "same as g5?",
        },
        {
            "instanceType": "g6e.2xlarge",
            "configurations": [
                {
                    "container": "TGI 2.2.0",
                    "quantization": "none",
                    "status": "OK",
                    "tokensPerSecond": "42.1",
                },
                {
                    "container": "SGLang 0.2.13",
                    "quantization": "none",
                    "status": "OK",
                    "tokensPerSecond": "45",
                },
                {
                    "container": "vLLM 0.5.5",
                    "quantization": "none",
                    "status": "OK",
                    "tokensPerSecond": "43.4",
                },
            ],
        },
        {
            "instanceType": "g6e.12xlarge",
            "configurations": [
                {
                    "container": "TGI 2.2.0",
                    "quantization": "none",
                    "status": "OK",
                    "tokensPerSecond": "112",
                },
                {
                    "container": "SGLang 0.2.13",
                    "quantization": "none",
                    "status": "OK",
                    "tokensPerSecond": "123",
                },
                {
                    "container": "vLLM 0.5.5",
                    "quantization": "none",
                    "status": "OK",
                    "tokensPerSecond": "106",
                },
            ],
        },
        {
            "instanceType": "p4d.24xlarge",
            "quantization": "none",
            "container": "TGI 2.2.0",
            "status": "OK",
            "tokensPerSecond": "145",
            "notes": '"MAX_INPUT_TOKENS": "40960", "MAX_TOTAL_TOKENS": "81920"\n\n64K/128K fails (even with 4-bit)',
        },
        {
            "instanceType": "inf2.*",
            "container": "TGI 2.2.0",
            "status": "not supported",
            "tokensPerSecond": "-",
            "notes": "Llama-3.1: TGI OK, Neuron SDK OK, optimum-neuron KO",
        },
        {
            "instanceType": "inf2.2xlarge",
            "container": "transformers-neuronx 0.11.351",
            "quantization": "none",
            "status": "OK",
            "tokensPerSecond": "24",
            "notes": "Neuron SDK 2.19.1",
        },
    ],
}