GB300-NVL72_GB300-288GB_aarch64x72_TRT
Submitted by Googleon 2026-06-26
Result statement
Google ran DeepSeek-R1 on GB300-NVL72_GB300-288GB_aarch64x72_TRT and achieved a verified 408,369.4 tokens/second System TPS at 4,171.9ms TTFT P90 in the Standardized division of the Available category .
[1] MLPerf Endpoints v1.0 DeepSeek-R1; retrieved from endpoints.mlcommons.org on October 22, 2026. Result —, run 9: concurrency = 8,192 queries, TTFT P90 = 4,171.9ms, System TPS = 408,369.4 tokens/sec and interactivity = 49.8 tokens/sec per user. Result verified by MLCommons Association. The MLPerf name and logo are registered and unregistered trademarks of MLCommons Association in the United States and other countries. All rights reserved. Unauthorized use strictly prohibited. See www.mlcommons.org for more information.
System TPS vs. Tokens/s per User
System TPS vs. Concurrency
Time to First Token P90 vs. Concurrency
Tokens/s per User vs. Concurrency
Processor
Accelerator
Host / Storage
Networking
Other
| Field Name | Run 1 | Run 2 | Run 3 | Run 4 | Run 5 | Run 6 | Run 7 | Run 8 | Run 9 |
|---|---|---|---|---|---|---|---|---|---|
| Run Date | 06/26/2026 | 06/26/2026 | 06/26/2026 | 06/26/2026 | 06/26/2026 | 06/26/2026 | 06/26/2026 | 06/26/2026 | 06/26/2026 |
| Concurrency | 4 | 48 | 64 | 256 | 512 | 1,024 | 2,048 | 4,096 | 8,192 |
| System Tokens/Second | 1,234.1 | 10,864.6 | 14,294.5 | 44,857.8 | 82,717.2 | 132,474.9 | 200,860.2 | 325,843.1 | 408,369.4 |
| Tokens/Second per User | 308.5 | 226.3 | 223.4 | 175.2 | 161.6 | 129.4 | 98.1 | 79.6 | 49.8 |
| TTFT P90 (ms) | 176.8 | 257.8 | 268.6 | 1943.7 | 297.3 | 308.5 | 1054.4 | 2064.9 | 4171.9 |
| Utilization | 0.3% | 2.7% | 3.5% | 11.0% | 20.3% | 32.4% | 49.2% | 79.8% | 100.0% |
| Configuration Summary | dsr1-gb300x72-pareto-nvl72-c4 · submission | dsr1-gb300x72-pareto-nvl72-c48 · submission | dsr1-gb300x72-pareto-nvl72-c64 · submission | dsr1-gb300x72-pareto-nvl72-c256 · submission | dsr1-gb300x72-pareto-nvl72-c512 · submission | dsr1-gb300x72-pareto-nvl72-c1024 · submission | dsr1-gb300x72-pareto-nvl72-c2048 · submission | dsr1-gb300x72-pareto-nvl72-c4096 · submission | dsr1-gb300x72-pareto-nvl72-c8192 · submission |
| Config Type | submission | submission | submission | submission | submission | submission | submission | submission | submission |
| Config Region | — | — | — | — | — | — | — | — | — |
| Target QPS | — | — | — | — | — | — | — | — | — |
| Result State | complete | complete | complete | complete | complete | complete | complete | complete | complete |
| Measured Accuracy Score | — | — | — | — | — | — | — | — | — |
| Benchmark Version | unknown | unknown | unknown | unknown | unknown | unknown | unknown | unknown | unknown |
| Result Version | 0.1.0 | 0.1.0 | 0.1.0 | 0.1.0 | 0.1.0 | 0.1.0 | 0.1.0 | 0.1.0 | 0.1.0 |
| Archive URI | gs://mlperf-endpoints-blob/runs/4c107b18-9410-4f80-90ce-88f9e75417d4/archive.tar.gz | gs://mlperf-endpoints-blob/runs/0d2b7b67-4ea7-43ee-9e37-54ff64cc5941/archive.tar.gz | gs://mlperf-endpoints-blob/runs/ee9d8ca8-ec42-4daa-b44b-8de26e089410/archive.tar.gz | gs://mlperf-endpoints-blob/runs/03c2178d-6e91-4a2f-b69e-5fae60578750/archive.tar.gz | gs://mlperf-endpoints-blob/runs/afd13761-b8ad-476d-9da5-ade649716249/archive.tar.gz | gs://mlperf-endpoints-blob/runs/06cccf32-a58a-447a-b76e-be5ff31e3f87/archive.tar.gz | gs://mlperf-endpoints-blob/runs/cecfc4d8-b47d-4190-9516-df3ee8f2591a/archive.tar.gz | gs://mlperf-endpoints-blob/runs/a817ab19-0489-41a5-b6ab-22745a8fb213/archive.tar.gz | gs://mlperf-endpoints-blob/runs/d17716a0-752a-4757-b484-4848c0038050/archive.tar.gz |
| QPS | 0.3296 | 2.9008 | 3.8139 | 11.9251 | 22.0091 | 35.1975 | 53.7055 | 88.7754 | 108.7729 |
| Total Output Tokens | 3,744,634 | 24,652,151 | 38,229,521 | 82,529,856 | 280,355,359 | 396,368,631 | 722,095,590 | 1,046,877,185 | 1,317,920,127 |
| Average Output Tokens | 3,744.63 | 3,745.39 | 3,747.99 | 3,761.62 | 3,758.32 | 3,763.76 | 3,740.03 | 3,670.42 | 3,754.33 |
| Run Duration (s) | 3,034.34 | 2,269.04 | 2,674.43 | 1,839.81 | 3,389.32 | 2,992.03 | 3,595.02 | 3,212.83 | 3,227.27 |
| Total Queries | 1,000 | 6,582 | 10,200 | 21,940 | 74,596 | 105,312 | 193,072 | 285,220 | 351,040 |
| Total Queries Issued | 1,000 | 6,582 | 10,200 | 21,940 | 74,596 | 105,312 | 193,072 | 285,220 | 351,040 |
| Total Queries Failed | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Time to First Token (TTFT) (ms) | |||||||||
| Minimum | 139.8 | 147.5 | 149.5 | 148.3 | 150.3 | 151.9 | 162.7 | 174.3 | 189.5 |
| Average | 206.6 | 199.1 | 214.8 | 758.1 | 328.0 | 368.3 | 497.2 | 838.3 | 1547.7 |
| P50 | 160.8 | 163.7 | 164.1 | 296.6 | 233.9 | 261.3 | 298.9 | 337.3 | 345.1 |
| P90 | 176.8 | 257.8 | 268.6 | 1943.7 | 297.3 | 308.5 | 1054.4 | 2064.9 | 4171.9 |
| P95 | 192.6 | 281.8 | 292.0 | 2465.6 | 309.2 | 322.6 | 1428.5 | 2908.7 | 6380.1 |
| P99 | 2136.2 | 337.3 | 511.7 | 3977.7 | 327.7 | 1284.4 | 2345.6 | 7045.2 | 18417.6 |
| P999 | 2371.8 | 2055.4 | 4166.0 | 4306.1 | 16029.1 | 15314.8 | 6665.3 | 15484.2 | 27085.2 |
| Maximum | 2436.1 | 2056.0 | 4167.0 | 4999.2 | 17199.2 | 15483.9 | 7667.5 | 16800.9 | 28647.5 |
| Time Per Output Token (TPOT) (ms) | |||||||||
| Minimum | 0.41 | 2.94 | 2.92 | 1.97 | 4.08 | 4.35 | 4.66 | 3.35 | 4.81 |
| Average | 3.28 | 4.39 | 4.45 | 5.42 | 6.12 | 7.62 | 10.0 | 12.2 | 19.1 |
| P50 | 3.27 | 4.34 | 4.37 | 5.37 | 6.05 | 7.55 | 9.94 | 12.1 | 19.0 |
| P90 | 3.77 | 4.85 | 4.90 | 6.01 | 6.77 | 8.46 | 11.2 | 13.5 | 21.5 |
| P95 | 3.95 | 4.95 | 5.06 | 6.12 | 6.90 | 8.62 | 11.4 | 13.8 | 21.9 |
| P99 | 4.20 | 5.24 | 6.02 | 6.34 | 7.14 | 8.92 | 11.8 | 14.3 | 22.7 |
| P999 | 4.36 | 5.81 | 7.47 | 6.62 | 7.44 | 9.27 | 12.3 | 14.9 | 23.6 |
| Maximum | 4.37 | 6.20 | 9.32 | 6.83 | 7.88 | 14.9 | 13.2 | 16.3 | 25.7 |
| Query Latency (ms) | |||||||||
| Minimum | 790.0 | 819.4 | 790.3 | 1060.3 | 1062.7 | 1323.9 | 1741.0 | 1967.6 | 2410.9 |
| Average | 12125.8 | 16222.8 | 16412.7 | 20639.6 | 22731.9 | 28290.6 | 36986.2 | 44431.7 | 71347.8 |
| P50 | 7189.0 | 9490.7 | 9368.6 | 12191.5 | 13182.7 | 16371.9 | 21314.7 | 26406.0 | 41386.9 |
| P90 | 28816.6 | 40134.1 | 40504.1 | 49976.5 | 55804.9 | 69551.4 | 91463.5 | 111636.5 | 175435.5 |
| P95 | 42836.3 | 56322.0 | 57573.2 | 70677.1 | 79668.1 | 99637.5 | 130027.8 | 158720.9 | 248289.1 |
| P99 | 59022.0 | 79539.8 | 80420.6 | 100511.6 | 112812.6 | 140778.9 | 180455.0 | 180980.1 | 354392.3 |
| P999 | 67050.1 | 87533.5 | 88910.9 | 109949.5 | 122736.7 | 153037.5 | 181135.6 | 181285.6 | 388915.5 |
| Maximum | 75437.5 | 93245.3 | 100102.6 | 134660.7 | 136603.6 | 180878.3 | 184593.9 | 196969.6 | 500784.6 |