Menu
Back previous/next report: k j

Intel 1-node-4x-BMG-Pro-B60

Submitted by Intelon 2026-02-27 Published on 2026-03-16

Submission Info
Submitter Intel
Submission ID intel-Intel-submission-1
Submission Status
Submission Date 2026-02-27
Publication Date 2026-03-16
Division Closed
Model
Model Name Llama-3.1-8B
Division Closed
Model Precision INT4
Calibration Link GitHub
Model Notes INT4 precision weights
Datasets
Dataset Name CNN_DailyMail
Dataset Type Accuracy + Performance
Average Input Tokens 870
Dataset Link Hugging Face

System TPS vs. Tokens/s per User

0204060System TPS0204060Tokens/s per User

System TPS vs. Concurrency

0204060System TPS05101520# of queries

Time to First Token P95 vs. Concurrency

0200400600800TTFT P95 (s)05101520# of queries

Tokens/s per User vs. Concurrency

0204060Tokens/s per User05101520# of queries
SUT Summary
Intel
System Name 1-node-4x-BMG-Pro-B60
System Availability Available
System Category Datacenter · —
System Size 4x Arc Pro B60
Node Types 1
Total Node Count 1
Max Supported Concurrency
Node Type ID 1
Node Count 1
Software
Ubuntu
Framework VLLM-0.11, vllm-xpu-kernels-0.0.1
Serving Framework VLLM-0.11, vllm-xpu-kernels-0.0.1
Inference Backend
Driver
Link to Container
Operating System Ubuntu 25.04
Filesystem EXT4
Other Software OneAPI® 2025.2
Software Notes
Hardware

Processor

Processor Model Name Intel® Xeon® w9-3475X Processor
Processors per Node 1
Cores Per Processor 36
VCPUs Per Processor 72

Accelerator

Accelerator Model Name Intel Arc Pro B60
Accelerators per Node 4
Memory Type GDDR6
Memory Capacity 24GB
Accelerator Interconnect
Host-Accelerator Interconnect PCIe Gen5 x8

Host / Storage

Host Memory Capacity 512GB
Memory Configuration 8x64GB DDR5 4800MTS
Storage Capacity 3.1TB
Storage Type SSD, EXT4 mounted disk storage

Networking

Network Interfaces 1, 1GB/s
Network Details Ethernet (on-board)
Network Topology

Other

Other Hardware
Cooling Air-cooled
Hardware Notes
Run Data
Field NameRun 1Run 2Run 3Run 4Run 5Run 6Run 7Run 8Run 9Run 10Run 11
Run Date02/25/202602/25/202602/25/202602/23/202602/23/202602/23/202602/23/202602/23/202602/23/202602/23/202602/23/2026
Concurrency0000011351021
System Tokens/Second1.32.65.09.316.728.542.255.063.567.267.0
Tokens/Second per User66.465.061.957.551.844.232.721.312.36.53.2
TTFT P95 (ms)11857.312090.412244.412547.413610.314835.718387.621752.530453.7100155.0867331.6
Utilization2.0%3.9%7.4%13.8%24.9%42.4%62.8%81.9%94.5%100.0%99.7%
Configuration SummaryDP=1 TP=1DP=2 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1DP=4 TP=1
Config Type
Config Region
Target QPS
Result State
Measured Accuracy Score
Benchmark Version
Result Version
Archive URI
QPS0.01050.02050.03900.07240.13060.22270.32970.42980.49610.52470.5231
Total Output Tokens1,711,1021,711,1021,711,1021,711,1021,711,1021,711,1021,711,1021,711,1021,711,1021,711,1021,711,102
Average Output Tokens
Run Duration (s)25,761.3913,163.166,914.983,722.972,064.731,210.81817.91627.34543.52513.86515.49
Total Queries13,36813,36813,36813,36813,36813,36813,36813,36813,36813,36813,368
Total Queries Issued
Total Queries Failed
Time to First Token (TTFT) P95 (ms)
Minimum1782.41853.61803.41959.22196.02182.12405.33201.54455.64261.775584.5
Average6000.16203.36297.06591.77066.77836.29890.712603.220014.644447.6481567.9
P505059.05505.15481.25753.96232.26788.38378.410342.714693.325190.5401338.9
P909574.19748.810045.810751.411755.812810.915607.718718.825459.266028.9793085.7
P9511857.312090.412244.412547.413610.314835.718387.621752.530453.7100155.0867331.6
P9914307.414568.214812.715241.017508.819781.025207.331961.5244075.7563573.61464105.9
P99915694.216382.817463.220669.669332.584593.6135089.9225352.4408852.8778556.31723303.9
Maximum61266.277066.179560.085803.790723.6120887.2159252.0278166.0434801.1829670.71810980.2
Time Per Output Token (TPOT) (ms)
Minimum692.4684.0697.7722.0761.8784.9849.4901.51088.21304.21920.0
Average705.0719.9758.0817.6908.41068.81448.22240.63893.87301.511299.4
P50704.5712.0752.8812.1902.21059.91437.52221.73845.86848.911605.9
P90708.5747.5793.5867.2982.31179.71654.72659.54864.110866.411703.5
P95709.7762.1809.7885.81008.41220.01718.42776.35104.711340.211727.9
P99724.2794.0839.4926.21062.41305.81842.22978.65587.811597.411778.0
P999770.4820.0870.8976.51125.11413.41987.93368.95823.811677.611836.7
Maximum829.0834.2902.41025.61205.81537.52081.83695.15931.211688.611857.6
Query Latency (ms)
Minimum89712.089785.291318.696279.5102115.6108817.9120585.7120932.0163055.8197397.3608648.4
Average95534.897627.0102558.9110423.0122435.9143572.6193813.8297157.2514529.1971738.61916585.8
P5094598.796893.8101739.2109680.7121563.7142274.4192135.8293809.3505315.7897076.11865249.5
P9099607.0103013.6108541.8117877.5132690.2158606.2220955.5351277.0641370.91450410.62250152.8
P95101839.4105195.0110774.5120776.8136558.4164342.0230063.6367688.8679688.31508837.82340533.1
P99104731.2109239.5115175.5126537.6143571.8176633.3250673.5400148.5754846.71601823.72921688.3
P999107338.6114554.5120879.0133681.6184259.1211439.8318532.2597134.8987869.51924953.93178950.8
Maximum151486.1169167.7175869.1209798.4208807.1258287.2340403.3632273.41027301.42013520.93261327.2