Skip to content

Linpack Benchmark Results

Compare Linpack floating-point performance across Single Board Computers

80 boards tested · scores from 2.66 to 156.4 (Performance (GFLOPS))

80 results

Full Results

Ranked by
0246
  1. 1
  2. 2
    Radxa Dragon Q6A 6GBQCS6490 · $119
    3.83
  3. 3
    3.52
  4. 4
    Raspberry Pi 500 8GBBCM2712 · $194
    3.41
  5. 5
    Radxa ROCK 4D 8GBRK3576 · $106
    3.27
  6. 6
    3.25
  7. 7
    Raspberry Pi 5 16GBBCM2712 · $329
    3.19
  8. 8
  9. 9
    Raspberry Pi 5 8GBBCM2712 · $189
    3.15
  10. 10
    3.12
  11. 11
    Fastest Overall: Radxa Dragon Q8B 32GBSnapdragon 8cx Gen 3 · $677
    3.1
  12. 12
    Orange Pi 5 Plus 8GBRK3588 · $135
    2.93
  13. 13
    Dshan Pi A1 4GBRK3576 · $69
    2.79
  14. 14
    Raspberry Pi 5 2GBBCM2712 · $82
    2.78
  15. 15
    Orange Pi 5 Pro 16GBRK3588S · $120
    2.76
  16. 16
    2.74
  17. 17
    Raspberry Pi 5 4GBBCM2712 · $118
    2.71
  18. 18
    2.67
  19. 19
    Raspberry Pi 500+ 16GBBCM2712 · $443
    2.66
  20. 20
    2.64
  21. 21
    Radxa ROCK 5A 4GBRK3588S · $130
    2.61
  22. 22
    ArmSoM Sige5 8GBRK3576 · $239
    2.61
  23. 23
    Radxa ROCK 5T 12GBRK3588 · $125
    2.54
  24. 24
    2.54
  25. 25
    2.53
  26. 26
    Fastest Under $100: Orange Pi 5B 4GBRK3588S · $75
    2.5
  27. 27
    ArmSoM AIM7 8GBRK3588 · $297
    2.49
  28. 28
    2.39
  29. 29
    Radxa ROCK 5B+ 16GBRK3588 · $135
    2.3
  30. 30
    Radxa NX5 8GBRK3588S · $119
    2.29
  31. 31
    ArmSoM W3 8GBRK3588 · $316
    2.27
  32. 32
    DEBIX Model A 4GBiMX 8M Plus · $135
    2.25
  33. 33
    Cool Pi 4B 4GBRK3588S · $140
    2.23
  34. 34
    DEBIX Model C 2GBiMX 93 · $65
    2.23
  35. 35
  36. 36
    Best Value: Radxa ROCK 2F 2GBRK3528A · $15
    2.21
  37. 37
    Banana Pi M5 4GBS905X3 · $146
    2.2
  38. 38
    Radxa CM5 Lite 8GBRK3582 · $119
    2.19
  39. 39
    Radxa ROCK 5 ITX 8GBRK3588 · $120
    2.17
  40. 40
    Radxa ROCK 5B 8GBRK3588 · $160
    2.17
  41. 41
    Banana Pi CM4 4GBA311D · $104
    2.15
  42. 42
    Orange Pi 4A 4GBT527 · $50
    2.1
  43. 43
    Radxa ROCK 5C 4GBRK3588S2 · $166
    2.1
  44. 44
  45. 45
    Radxa ROCK 2A 4GBRK3528A · $30
    2.08
  46. 46
    LattePanda Mu 8GBN100 · $398
    2.06
  47. 47
    Forlinx FET3576-C 2GBRK3576 · $210
    2.05
  48. 48
    2.04
  49. 49
    Banana Pi R4 8GBMT7988A · $146
    2.01
  50. 50
    Raspberry Pi 400 4GBBCM2711 · $65
    2
  51. 51
    1.99
  52. 52
    ArmSoM Sige7 8GBRK3588 · $312
    1.99
  53. 53
    ArmSoM Sige1 2GBRK3528A · $48
    1.97
  54. 54
    1.93
  55. 55
    Khadas VIM1S 2GBS905Y4 · $59
    1.89
  56. 56
    Raspberry Pi 4 2GBBCM2711 · $72
    1.83
  57. 57
    Radxa Zero 3W 8GBRK3566 · $92
    1.75
  58. 58
  59. 59
    Orange Pi CM4 2GBRK3566 · $40
    1.41
  60. 60
    Radxa ROCK 3C 2GBRK3566 · $40
    1.37
  61. 61
  62. 62
    Orange Pi 3B 4GBRK3566 · $94
    1.27
  63. 63
    Radxa X4 8GBN100 · $100
    1.23
  64. 64
    NVIDIA Jetson Nano 2GBTegra X1 · $49
    1.13
  65. 65
    Radxa ROCK 3B 2GBRK3568 · $51
    1.01
  66. 66
    Radxa 4 SE 4GBRK3399-T · $80
    1
  67. 67
    Radxa ROCK 4C+ 4GBRK3399-T · $80
    0.99
  68. 68
    Radxa ROCK 4B+ 4GBRK3399 OP1 · $170
    0.92
  69. 69
    Radxa CM3I 2GBRK3568 · $65
    0.83
  70. 70
    0.78
  71. 71
    Milk-V Mars 4GBJH7110 · $49
    0.69
  72. 72
    0.58
  73. 73
    Orange Pi RV 4GBJH7110 · $50
    0.49
  74. 74
    0.47
0246
GFLOPS/W, higher is better
Showing 74 boards · median 2.2 (6 without a GFLOPS/W result not shown)Top Picks from above

Frequently Asked Questions

What is the fastest single board computer in Linpack?

The Radxa Dragon Q8B 32GB currently tops our Linpack ranking with 156.4 (Performance (GFLOPS)), measured on real hardware with our standardised test setup.

What is a good Linpack score for an SBC?

Across the 80 boards we have tested, Linpack results range from 2.66 to 156.4 (Performance (GFLOPS)), with a median of 21.5. Anything above the median puts a board in the faster half of the market.

About Linpack testing

Linpack measures raw floating point throughput in GFLOPS by solving dense systems of linear equations, the same family of workload used to rank the world's supercomputers. It leans hard on the CPU's vector units and memory subsystem, making it a strong indicator for scientific computing, signal processing, and other maths-heavy workloads. Where we have power measurements, we also report GFLOPS per watt, which turns Linpack into one of the best efficiency comparisons between boards.