Oracle Cloud Infrastructure Documentation

Scenario 2: Retrieval-Augmented Generation (RAG) Benchmarks in Generative AI

The RAG scenario has a very long prompt and a short response. This scenario also mimics summarization use cases.

  • The prompt length is fixed to 2,000 tokens.
  • The response length is fixed to 200 tokens.
Important

The performance (inference speed, throughput, latency) of a hosting dedicated AI cluster depends on the traffic scenarios going through the model that it's hosting. Traffic scenarios depend on:

  1. The number of concurrent requests.
  2. The number of tokens in the prompt.
  3. The number of tokens in the response.
  4. The variance of (2) and (3) across requests.

Review the terms used in the hosting dedicated AI cluster benchmarks. For a list of scenarios and their descriptions, see Chat and Text Generation Scenarios. The retrieval-augmented generation scenario is performed in the following region.

Brazil East (Sao Paulo)

Model: cohere.command-r-08-2024 (Cohere Command R 08-2024) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 136.91 133.84 3.19 18.35
2 128.58 250.14 3.40 34.21
4 114.22 434.70 3.81 59.56
8 93.74 680.93 4.67 92.38
16 71.06 1,007.40 5.96 138.94
32 50.30 1,561.75 8.74 212.91
64 30.71 1,922.54 14.28 262.99
128 17.99 2,043.92 25.57 279.72
256 8.83 2,061.45 46.83 281.73
Model: cohere.command-r-plus-08-2024 (Cohere Command R+ 08-2024) model hosted on one Large Cohere V2_2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.30 103.49 4.27 13.81
2 99.67 195.23 4.51 26.15
4 92.17 349.80 4.87 46.61
8 73.11 532.86 6.08 71.15
16 54.17 750.15 6.08 99.56
32 40.22 1,266.6 11.18 169.29
64 24.62 1,559.01 18.31 208.03
128 15.35 1,604.24 31.44 213.95
256 6.96 1,660.81 58.06 221.39
Model: meta.llama-3.3-70b-instruct (Meta Llama 3.3 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 53.37 50.55 3.93 15.16
2 51.06 96.43 4.11 28.93
4 48.31 183.35 4.33 55.01
8 41.85 309.97 5.05 92.99
16 33.93 489.93 6.29 146.98
32 23.61 663.30 9.12 198.99
64 18.27 724.11 15.87 217.23
128 18.27 724.07 28.49 217.22
256 18.35 709.78 45.89 212.94
Model: meta.llama-3.2-90b-vision-instruct (Meta Llama 3.2 90B Vision) model (text input only) hosted on one Large Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.83 44.33 4.47 13.30
2 46.14 82.67 4.79 24.80
4 45.18 145.33 5.46 43.60
8 44.67 234.67 6.74 70.40
16 43.43 336.00 9.34 100.80
32 32.74 394.66 15.61 118.40
64 33.25 416.00 30.12 124.80
128 33.28 405.32 59.98 121.60
256 33.27 394.60 116.63 118.38
Model: meta.llama-3.2-11b-vision-instruct (Meta Llama 3.2 11B Vision) model (text input only) hosted on one Small Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.23 101.67 1.95 30.50
2 100.86 191.33 2.08 57.40
4 96.79 348.00 2.28 104.40
8 86.60 568.00 2.77 170.40
16 72.41 837.33 3.73 251.20
32 43.23 1,002.66 6.19 300.80
64 47.43 1,066.65 11.63 320.00
128 47.45 1,066.62 23.25 319.99
256 47.41 1,066.60 45.83 319.98
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 2 unit of a dedicated AI cluster
Important

You can host the meta.llama-3.1-405b-instruct model only on a dedicated AI cluster of type Large Generic 2. This type in intended to provide better throughput with less hardware and a lower cost than its predecessor, Large Generic 4.

The following tables provide benchmarks that were performed for the meta.llama-3.1-405b-instruct model hosted on one Large Generic 2 unit and on one Large Generic 4 unit. If your model is currently hosted on a Large Generic 4 unit, compare the following tables to decide whether to host the model on this new unit.

Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 27.30 25.65 7.74 7.69
2 25.70 48.30 8.21 14.49
4 23.48 88.27 8.96 26.48
8 20.09 150.57 10.52 45.17
16 14.89 223.85 14.10 67.15
32 10.97 330.10 19.10 99.03
64 8.80 386.54 32.06 115.96
128 8.82 386.74 62.04 116.02
256 8.82 375.21 119.99 112.56
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 4 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 32.94 25.28 7.91 7.58
2 31.31 49.05 8.15 14.71
4 28.85 87.28 8.85 26.18
8 24.24 141.04 10.42 42.31
16 20.31 219.48 12.52 65.85
32 15.99 366.75 16.70 110.03
64 11.03 485.78 24.63 145.74
128 8.27 560.24 41.22 168.07
256 8.01 583.97 74.21 175.19
Model: meta.llama-3.1-70b-instruct (Meta Llama 3.1 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 95.86 49.82 4.10 14.62
2 91.14 94.21 4.34 14.62
4 84.77 170.89 4.63 50.04
8 75.09 281.23 5.35 82.35
16 58.20 407.94 7.00 82.35
32 42.16 593.60 10.26 174.28
64 31.93 715.30 16.44 174.28
128 30.32 754.79 29.37 174.28
256 29.16 751.22 56.21 220.34
Model: meta.llama-3-70b-instruct (Meta Llama 3) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.78 47.82 4.28 14.02
2 45.51 90.14 4.50 26.42
4 42.24 164.92 4.81 48.51
8 37.44 289.82 5.48 85.13
16 28.00 421.00 7.19 123.72
32 18.73 542.99 10.65 159.56
64 11.63 668.78 16.17 196.44
128 6.20 700.83 32.89 205.70
256 3.97 756.00 54.71 222.02
Model: cohere.command-r-16k (Cohere Command R) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 49.33 47.66 4.14 14.24
2 45.65 86.90 4.50 26.04
4 40.32 152.10 5.09 45.51
8 30.69 235.78 6.57 70.43
16 24.60 310.44 9.74 93.07
32 9.95 307.32 18.21 91.81
64 5.43 297.06 31.41 89.08
128 4.44 313.47 44.90 93.89
256 2.36 312.97 85.35 93.53
Model: cohere.command-r-plus (Cohere Command R+) model hosted on one Large Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 107.17 94.33 4.17 14.12
2 100.71 176.04 4.44 26.35
4 90.03 310.18 4.96 46.44
8 70.71 493.30 6.26 73.86
16 53.45 716.66 8.20 108.07
32 35.60 929.63 12.22 139.13
64 21.75 1,150.16 18.41 172.14
128 17.99 1,209.36 31.93 181.05
256 9.19 1,213.82 53.31 181.70

Germany Central (Frankfurt)

Model: cohere.command-r-08-2024 (Cohere Command R 08-2024) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 136.91 133.84 3.19 18.35
2 128.58 250.14 3.40 34.21
4 114.22 434.70 3.81 59.56
8 93.74 680.93 4.67 92.38
16 71.06 1,007.40 5.96 138.94
32 50.30 1,561.75 8.74 212.91
64 30.71 1,922.54 14.28 262.99
128 17.99 2,043.92 25.57 279.72
256 8.83 2,061.45 46.83 281.73
Model: cohere.command-r-plus-08-2024 (Cohere Command R+ 08-2024) model hosted on one Large Cohere V2_2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.30 103.49 4.27 13.81
2 99.67 195.23 4.51 26.15
4 92.17 349.80 4.87 46.61
8 73.11 532.86 6.08 71.15
16 54.17 750.15 6.08 99.56
32 40.22 1,266.6 11.18 169.29
64 24.62 1,559.01 18.31 208.03
128 15.35 1,604.24 31.44 213.95
256 6.96 1,660.81 58.06 221.39
Model: meta.llama-3.3-70b-instruct (Meta Llama 3.3 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 53.37 50.55 3.93 15.16
2 51.06 96.43 4.11 28.93
4 48.31 183.35 4.33 55.01
8 41.85 309.97 5.05 92.99
16 33.93 489.93 6.29 146.98
32 23.61 663.30 9.12 198.99
64 18.27 724.11 15.87 217.23
128 18.27 724.07 28.49 217.22
256 18.35 709.78 45.89 212.94
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 2 unit of a dedicated AI cluster
Important

You can host the meta.llama-3.1-405b-instruct model only on a dedicated AI cluster of type Large Generic 2. This type in intended to provide better throughput with less hardware and a lower cost than its predecessor, Large Generic 4.

The following tables provide benchmarks that were performed for the meta.llama-3.1-405b-instruct model hosted on one Large Generic 2 unit and on one Large Generic 4 unit. If your model is currently hosted on a Large Generic 4 unit, compare the following tables to decide whether to host the model on this new unit.

Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 27.30 25.65 7.74 7.69
2 25.70 48.30 8.21 14.49
4 23.48 88.27 8.96 26.48
8 20.09 150.57 10.52 45.17
16 14.89 223.85 14.10 67.15
32 10.97 330.10 19.10 99.03
64 8.80 386.54 32.06 115.96
128 8.82 386.74 62.04 116.02
256 8.82 375.21 119.99 112.56
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 4 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 32.94 25.28 7.91 7.58
2 31.31 49.05 8.15 14.71
4 28.85 87.28 8.85 26.18
8 24.24 141.04 10.42 42.31
16 20.31 219.48 12.52 65.85
32 15.99 366.75 16.70 110.03
64 11.03 485.78 24.63 145.74
128 8.27 560.24 41.22 168.07
256 8.01 583.97 74.21 175.19
Model: meta.llama-3.1-70b-instruct (Meta Llama 3.1 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 95.86 49.82 4.10 14.62
2 91.14 94.21 4.34 14.62
4 84.77 170.89 4.63 50.04
8 75.09 281.23 5.35 82.35
16 58.20 407.94 7.00 82.35
32 42.16 593.60 10.26 174.28
64 31.93 715.30 16.44 174.28
128 30.32 754.79 29.37 174.28
256 29.16 751.22 56.21 220.34
Model: meta.llama-3-70b-instruct (Meta Llama 3) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.78 47.82 4.28 14.02
2 45.51 90.14 4.50 26.42
4 42.24 164.92 4.81 48.51
8 37.44 289.82 5.48 85.13
16 28.00 421.00 7.19 123.72
32 18.73 542.99 10.65 159.56
64 11.63 668.78 16.17 196.44
128 6.20 700.83 32.89 205.70
256 3.97 756.00 54.71 222.02
Model: cohere.command-r-16k (Cohere Command R) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 49.33 47.66 4.14 14.24
2 45.65 86.90 4.50 26.04
4 40.32 152.10 5.09 45.51
8 30.69 235.78 6.57 70.43
16 24.60 310.44 9.74 93.07
32 9.95 307.32 18.21 91.81
64 5.43 297.06 31.41 89.08
128 4.44 313.47 44.90 93.89
256 2.36 312.97 85.35 93.53
Model: cohere.command-r-plus (Cohere Command R+) model hosted on one Large Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 107.17 94.33 4.17 14.12
2 100.71 176.04 4.44 26.35
4 90.03 310.18 4.96 46.44
8 70.71 493.30 6.26 73.86
16 53.45 716.66 8.20 108.07
32 35.60 929.63 12.22 139.13
64 21.75 1,150.16 18.41 172.14
128 17.99 1,209.36 31.93 181.05
256 9.19 1,213.82 53.31 181.70

Japan Central (Osaka)

Model: cohere.command-r-08-2024 (Cohere Command R 08-2024) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 136.91 133.84 3.19 18.35
2 128.58 250.14 3.40 34.21
4 114.22 434.70 3.81 59.56
8 93.74 680.93 4.67 92.38
16 71.06 1,007.40 5.96 138.94
32 50.30 1,561.75 8.74 212.91
64 30.71 1,922.54 14.28 262.99
128 17.99 2,043.92 25.57 279.72
256 8.83 2,061.45 46.83 281.73
Model: cohere.command-r-plus-08-2024 (Cohere Command R+ 08-2024) model hosted on one Large Cohere V2_2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.30 103.49 4.27 13.81
2 99.67 195.23 4.51 26.15
4 92.17 349.80 4.87 46.61
8 73.11 532.86 6.08 71.15
16 54.17 750.15 6.08 99.56
32 40.22 1,266.6 11.18 169.29
64 24.62 1,559.01 18.31 208.03
128 15.35 1,604.24 31.44 213.95
256 6.96 1,660.81 58.06 221.39
Model: meta.llama-3.3-70b-instruct (Meta Llama 3.3 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 53.37 50.55 3.93 15.16
2 51.06 96.43 4.11 28.93
4 48.31 183.35 4.33 55.01
8 41.85 309.97 5.05 92.99
16 33.93 489.93 6.29 146.98
32 23.61 663.30 9.12 198.99
64 18.27 724.11 15.87 217.23
128 18.27 724.07 28.49 217.22
256 18.35 709.78 45.89 212.94
Model: meta.llama-3.2-90b-vision-instruct (Meta Llama 3.2 90B Vision) model (text input only) hosted on one Large Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.83 44.33 4.47 13.30
2 46.14 82.67 4.79 24.80
4 45.18 145.33 5.46 43.60
8 44.67 234.67 6.74 70.40
16 43.43 336.00 9.34 100.80
32 32.74 394.66 15.61 118.40
64 33.25 416.00 30.12 124.80
128 33.28 405.32 59.98 121.60
256 33.27 394.60 116.63 118.38
Model: meta.llama-3.2-11b-vision-instruct (Meta Llama 3.2 11B Vision) model (text input only) hosted on one Small Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.23 101.67 1.95 30.50
2 100.86 191.33 2.08 57.40
4 96.79 348.00 2.28 104.40
8 86.60 568.00 2.77 170.40
16 72.41 837.33 3.73 251.20
32 43.23 1,002.66 6.19 300.80
64 47.43 1,066.65 11.63 320.00
128 47.45 1,066.62 23.25 319.99
256 47.41 1,066.60 45.83 319.98
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 2 unit of a dedicated AI cluster
Important

You can host the meta.llama-3.1-405b-instruct model only on a dedicated AI cluster of type Large Generic 2. This type in intended to provide better throughput with less hardware and a lower cost than its predecessor, Large Generic 4.

The following tables provide benchmarks that were performed for the meta.llama-3.1-405b-instruct model hosted on one Large Generic 2 unit and on one Large Generic 4 unit. If your model is currently hosted on a Large Generic 4 unit, compare the following tables to decide whether to host the model on this new unit.

Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 27.30 25.65 7.74 7.69
2 25.70 48.30 8.21 14.49
4 23.48 88.27 8.96 26.48
8 20.09 150.57 10.52 45.17
16 14.89 223.85 14.10 67.15
32 10.97 330.10 19.10 99.03
64 8.80 386.54 32.06 115.96
128 8.82 386.74 62.04 116.02
256 8.82 375.21 119.99 112.56
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 4 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 32.94 25.28 7.91 7.58
2 31.31 49.05 8.15 14.71
4 28.85 87.28 8.85 26.18
8 24.24 141.04 10.42 42.31
16 20.31 219.48 12.52 65.85
32 15.99 366.75 16.70 110.03
64 11.03 485.78 24.63 145.74
128 8.27 560.24 41.22 168.07
256 8.01 583.97 74.21 175.19
Model: meta.llama-3.1-70b-instruct (Meta Llama 3.1 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 95.86 49.82 4.10 14.62
2 91.14 94.21 4.34 14.62
4 84.77 170.89 4.63 50.04
8 75.09 281.23 5.35 82.35
16 58.20 407.94 7.00 82.35
32 42.16 593.60 10.26 174.28
64 31.93 715.30 16.44 174.28
128 30.32 754.79 29.37 174.28
256 29.16 751.22 56.21 220.34
Model: cohere.command-r-16k (Cohere Command R) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 49.33 47.66 4.14 14.24
2 45.65 86.90 4.50 26.04
4 40.32 152.10 5.09 45.51
8 30.69 235.78 6.57 70.43
16 24.60 310.44 9.74 93.07
32 9.95 307.32 18.21 91.81
64 5.43 297.06 31.41 89.08
128 4.44 313.47 44.90 93.89
256 2.36 312.97 85.35 93.53
Model: cohere.command-r-plus (Cohere Command R+) model hosted on one Large Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 107.17 94.33 4.17 14.12
2 100.71 176.04 4.44 26.35
4 90.03 310.18 4.96 46.44
8 70.71 493.30 6.26 73.86
16 53.45 716.66 8.20 108.07
32 35.60 929.63 12.22 139.13
64 21.75 1,150.16 18.41 172.14
128 17.99 1,209.36 31.93 181.05
256 9.19 1,213.82 53.31 181.70

UK South (London)

Model: cohere.command-r-08-2024 (Cohere Command R 08-2024) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 136.91 133.84 3.19 18.35
2 128.58 250.14 3.40 34.21
4 114.22 434.70 3.81 59.56
8 93.74 680.93 4.67 92.38
16 71.06 1,007.40 5.96 138.94
32 50.30 1,561.75 8.74 212.91
64 30.71 1,922.54 14.28 262.99
128 17.99 2,043.92 25.57 279.72
256 8.83 2,061.45 46.83 281.73
Model: cohere.command-r-plus-08-2024 (Cohere Command R+ 08-2024) model hosted on one Large Cohere V2_2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.30 103.49 4.27 13.81
2 99.67 195.23 4.51 26.15
4 92.17 349.80 4.87 46.61
8 73.11 532.86 6.08 71.15
16 54.17 750.15 6.08 99.56
32 40.22 1,266.6 11.18 169.29
64 24.62 1,559.01 18.31 208.03
128 15.35 1,604.24 31.44 213.95
256 6.96 1,660.81 58.06 221.39
Model: meta.llama-3.3-70b-instruct (Meta Llama 3.3 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 53.37 50.55 3.93 15.16
2 51.06 96.43 4.11 28.93
4 48.31 183.35 4.33 55.01
8 41.85 309.97 5.05 92.99
16 33.93 489.93 6.29 146.98
32 23.61 663.30 9.12 198.99
64 18.27 724.11 15.87 217.23
128 18.27 724.07 28.49 217.22
256 18.35 709.78 45.89 212.94
Model: meta.llama-3.2-90b-vision-instruct (Meta Llama 3.2 90B Vision) model (text input only) hosted on one Large Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.83 44.33 4.47 13.30
2 46.14 82.67 4.79 24.80
4 45.18 145.33 5.46 43.60
8 44.67 234.67 6.74 70.40
16 43.43 336.00 9.34 100.80
32 32.74 394.66 15.61 118.40
64 33.25 416.00 30.12 124.80
128 33.28 405.32 59.98 121.60
256 33.27 394.60 116.63 118.38
Model: meta.llama-3.2-11b-vision-instruct (Meta Llama 3.2 11B Vision) model (text input only) hosted on one Small Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.23 101.67 1.95 30.50
2 100.86 191.33 2.08 57.40
4 96.79 348.00 2.28 104.40
8 86.60 568.00 2.77 170.40
16 72.41 837.33 3.73 251.20
32 43.23 1,002.66 6.19 300.80
64 47.43 1,066.65 11.63 320.00
128 47.45 1,066.62 23.25 319.99
256 47.41 1,066.60 45.83 319.98
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 2 unit of a dedicated AI cluster
Important

You can host the meta.llama-3.1-405b-instruct model only on a dedicated AI cluster of type Large Generic 2. This type in intended to provide better throughput with less hardware and a lower cost than its predecessor, Large Generic 4.

The following tables provide benchmarks that were performed for the meta.llama-3.1-405b-instruct model hosted on one Large Generic 2 unit and on one Large Generic 4 unit. If your model is currently hosted on a Large Generic 4 unit, compare the following tables to decide whether to host the model on this new unit.

Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 27.30 25.65 7.74 7.69
2 25.70 48.30 8.21 14.49
4 23.48 88.27 8.96 26.48
8 20.09 150.57 10.52 45.17
16 14.89 223.85 14.10 67.15
32 10.97 330.10 19.10 99.03
64 8.80 386.54 32.06 115.96
128 8.82 386.74 62.04 116.02
256 8.82 375.21 119.99 112.56
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 4 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 32.94 25.28 7.91 7.58
2 31.31 49.05 8.15 14.71
4 28.85 87.28 8.85 26.18
8 24.24 141.04 10.42 42.31
16 20.31 219.48 12.52 65.85
32 15.99 366.75 16.70 110.03
64 11.03 485.78 24.63 145.74
128 8.27 560.24 41.22 168.07
256 8.01 583.97 74.21 175.19
Model: meta.llama-3.1-70b-instruct (Meta Llama 3.1 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 95.86 49.82 4.10 14.62
2 91.14 94.21 4.34 14.62
4 84.77 170.89 4.63 50.04
8 75.09 281.23 5.35 82.35
16 58.20 407.94 7.00 82.35
32 42.16 593.60 10.26 174.28
64 31.93 715.30 16.44 174.28
128 30.32 754.79 29.37 174.28
256 29.16 751.22 56.21 220.34
Model: meta.llama-3-70b-instruct (Meta Llama 3) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.78 47.82 4.28 14.02
2 45.51 90.14 4.50 26.42
4 42.24 164.92 4.81 48.51
8 37.44 289.82 5.48 85.13
16 28.00 421.00 7.19 123.72
32 18.73 542.99 10.65 159.56
64 11.63 668.78 16.17 196.44
128 6.20 700.83 32.89 205.70
256 3.97 756.00 54.71 222.02
Model: cohere.command-r-16k (Cohere Command R) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 49.33 47.66 4.14 14.24
2 45.65 86.90 4.50 26.04
4 40.32 152.10 5.09 45.51
8 30.69 235.78 6.57 70.43
16 24.60 310.44 9.74 93.07
32 9.95 307.32 18.21 91.81
64 5.43 297.06 31.41 89.08
128 4.44 313.47 44.90 93.89
256 2.36 312.97 85.35 93.53
Model: cohere.command-r-plus (Cohere Command R+) model hosted on one Large Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 107.17 94.33 4.17 14.12
2 100.71 176.04 4.44 26.35
4 90.03 310.18 4.96 46.44
8 70.71 493.30 6.26 73.86
16 53.45 716.66 8.20 108.07
32 35.60 929.63 12.22 139.13
64 21.75 1,150.16 18.41 172.14
128 17.99 1,209.36 31.93 181.05
256 9.19 1,213.82 53.31 181.70

US Midwest (Chicago)

Model: cohere.command-r-08-2024 (Cohere Command R 08-2024) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 136.91 133.84 3.19 18.35
2 128.58 250.14 3.40 34.21
4 114.22 434.70 3.81 59.56
8 93.74 680.93 4.67 92.38
16 71.06 1,007.40 5.96 138.94
32 50.30 1,561.75 8.74 212.91
64 30.71 1,922.54 14.28 262.99
128 17.99 2,043.92 25.57 279.72
256 8.83 2,061.45 46.83 281.73
Model: cohere.command-r-plus-08-2024 (Cohere Command R+ 08-2024) model hosted on one Large Cohere V2_2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.30 103.49 4.27 13.81
2 99.67 195.23 4.51 26.15
4 92.17 349.80 4.87 46.61
8 73.11 532.86 6.08 71.15
16 54.17 750.15 6.08 99.56
32 40.22 1,266.6 11.18 169.29
64 24.62 1,559.01 18.31 208.03
128 15.35 1,604.24 31.44 213.95
256 6.96 1,660.81 58.06 221.39
Model: meta.llama-3.3-70b-instruct (Meta Llama 3.3 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 53.37 50.55 3.93 15.16
2 51.06 96.43 4.11 28.93
4 48.31 183.35 4.33 55.01
8 41.85 309.97 5.05 92.99
16 33.93 489.93 6.29 146.98
32 23.61 663.30 9.12 198.99
64 18.27 724.11 15.87 217.23
128 18.27 724.07 28.49 217.22
256 18.35 709.78 45.89 212.94
Model: meta.llama-3.2-90b-vision-instruct (Meta Llama 3.2 90B Vision) model (text input only) hosted on one Large Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 47.83 44.33 4.47 13.30
2 46.14 82.67 4.79 24.80
4 45.18 145.33 5.46 43.60
8 44.67 234.67 6.74 70.40
16 43.43 336.00 9.34 100.80
32 32.74 394.66 15.61 118.40
64 33.25 416.00 30.12 124.80
128 33.28 405.32 59.98 121.60
256 33.27 394.60 116.63 118.38
Model: meta.llama-3.2-11b-vision-instruct (Meta Llama 3.2 11B Vision) model (text input only) hosted on one Small Generic V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 105.23 101.67 1.95 30.50
2 100.86 191.33 2.08 57.40
4 96.79 348.00 2.28 104.40
8 86.60 568.00 2.77 170.40
16 72.41 837.33 3.73 251.20
32 43.23 1,002.66 6.19 300.80
64 47.43 1,066.65 11.63 320.00
128 47.45 1,066.62 23.25 319.99
256 47.41 1,066.60 45.83 319.98
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 2 unit of a dedicated AI cluster
Important

You can host the meta.llama-3.1-405b-instruct model only on a dedicated AI cluster of type Large Generic 2. This type in intended to provide better throughput with less hardware and a lower cost than its predecessor, Large Generic 4.

The following tables provide benchmarks that were performed for the meta.llama-3.1-405b-instruct model hosted on one Large Generic 2 unit and on one Large Generic 4 unit. If your model is currently hosted on a Large Generic 4 unit, compare the following tables to decide whether to host the model on this new unit.

Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 27.30 25.65 7.74 7.69
2 25.70 48.30 8.21 14.49
4 23.48 88.27 8.96 26.48
8 20.09 150.57 10.52 45.17
16 14.89 223.85 14.10 67.15
32 10.97 330.10 19.10 99.03
64 8.80 386.54 32.06 115.96
128 8.82 386.74 62.04 116.02
256 8.82 375.21 119.99 112.56
Model: meta.llama-3.1-405b-instruct (Meta Llama 3.1 (405B)) model hosted on one Large Generic 4 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 32.94 25.28 7.91 7.58
2 31.31 49.05 8.15 14.71
4 28.85 87.28 8.85 26.18
8 24.24 141.04 10.42 42.31
16 20.31 219.48 12.52 65.85
32 15.99 366.75 16.70 110.03
64 11.03 485.78 24.63 145.74
128 8.27 560.24 41.22 168.07
256 8.01 583.97 74.21 175.19
Model: meta.llama-3.1-70b-instruct (Meta Llama 3.1 (70B)) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 95.86 49.82 4.10 14.62
2 91.14 94.21 4.34 14.62
4 84.77 170.89 4.63 50.04
8 75.09 281.23 5.35 82.35
16 58.20 407.94 7.00 82.35
32 42.16 593.60 10.26 174.28
64 31.93 715.30 16.44 174.28
128 30.32 754.79 29.37 174.28
256 29.16 751.22 56.21 220.34
Model: meta.llama-3-70b-instruct (Meta Llama 3) model hosted on one Large Generic unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 28.84 28.82 7.11 8.44
2 26.52 52.69 7.66 15.51
4 24.23 94.86 8.38 27.92
8 20.01 155.97 10.21 45.76
16 14.34 216.26 14.12 63.43
32 9.33 275.28 21.30 80.89
64 5.68 334.46 32.55 98.11
128 3.13 364.18 64.59 106.94
256 1.59 359.21 128.67 105.44
Model: cohere.command-r-16k (Cohere Command R) model hosted on one Small Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 49.33 47.66 4.14 14.24
2 45.65 86.90 4.50 26.04
4 40.32 152.10 5.09 45.51
8 30.69 235.78 6.57 70.43
16 24.60 310.44 9.74 93.07
32 9.95 307.32 18.21 91.81
64 5.43 297.06 31.41 89.08
128 4.44 313.47 44.90 93.89
256 2.36 312.97 85.35 93.53
Model: cohere.command-r-plus (Cohere Command R+) model hosted on one Large Cohere V2 unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 107.17 94.33 4.17 14.12
2 100.71 176.04 4.44 26.35
4 90.03 310.18 4.96 46.44
8 70.71 493.30 6.26 73.86
16 53.45 716.66 8.20 108.07
32 35.60 929.63 12.22 139.13
64 21.75 1,150.16 18.41 172.14
128 17.99 1,209.36 31.93 181.05
256 9.19 1,213.82 53.31 181.70
Model: cohere.command (Cohere Command 52 B) model hosted on one Large Cohere unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 33.13 25.28 6.68 8.62
8 23.24 90.64 13.29 29.84
32 13.03 163.48 26.56 54.21
128 5.60 186.31 65.30 61.32
Model: cohere.command-light (Cohere Command Light 6 B) model hosted on one Small Cohere unit of a dedicated AI cluster
Concurrency Token-level Inference Speed (token/second) Token-level Throughput (token/second) Request-level Latency (second) Request-level Throughput (Request per minute) (RPM)
1 56.71 50.88 3.14 17.61
8 24.70 148.42 6.15 53.93
32 11.06 235.31 13.37 85.14
128 3.40 280.3 31.64 105.77