Guest User

Untitled

a guest
Aug 27th, 2026
51
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 92.16 KB | None | 0 0
  1. Testing 4 devices
  2.  
  3. Backend 1/4: ROCm0
  4. Device description: AMD Radeon Graphics
  5. Device memory: 32752 MB (32732 MB free)
  6.  
  7. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 12780 runs - 83.26 us/run - 117.44 MFLOP/run -  1.41 TFLOPS
  8. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 12780 runs - 78.29 us/run - 117.44 MFLOP/run -  1.50 TFLOPS
  9. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 10224 runs - 102.54 us/run - 117.44 MFLOP/run -  1.15 TFLOPS
  10. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11928 runs - 88.25 us/run - 117.44 MFLOP/run -  1.33 TFLOPS
  11. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9372 runs - 110.90 us/run - 117.44 MFLOP/run -  1.06 TFLOPS
  12. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2556 runs - 443.49 us/run - 117.44 MFLOP/run - 264.81 GFLOPS
  13. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9372 runs - 109.78 us/run - 117.44 MFLOP/run -  1.07 TFLOPS
  14. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11928 runs - 84.48 us/run - 117.44 MFLOP/run -  1.39 TFLOPS
  15. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 14484 runs - 70.76 us/run - 117.44 MFLOP/run -  1.66 TFLOPS
  16. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 13632 runs - 76.87 us/run - 117.44 MFLOP/run -  1.53 TFLOPS
  17. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 10224 runs - 102.90 us/run - 117.44 MFLOP/run -  1.14 TFLOPS
  18. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 12780 runs - 81.11 us/run - 117.44 MFLOP/run -  1.45 TFLOPS
  19. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11076 runs - 94.06 us/run - 117.44 MFLOP/run -  1.25 TFLOPS
  20. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9372 runs - 113.82 us/run - 117.44 MFLOP/run -  1.03 TFLOPS
  21. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8520 runs - 127.10 us/run - 117.44 MFLOP/run - 923.99 GFLOPS
  22. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8520 runs - 129.68 us/run - 117.44 MFLOP/run - 905.60 GFLOPS
  23. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7668 runs - 135.63 us/run - 117.44 MFLOP/run - 865.90 GFLOPS
  24. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6816 runs - 159.90 us/run - 117.44 MFLOP/run - 734.45 GFLOPS
  25. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9372 runs - 114.56 us/run - 117.44 MFLOP/run -  1.03 TFLOPS
  26. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11928 runs - 88.38 us/run - 117.44 MFLOP/run -  1.33 TFLOPS
  27. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6816 runs - 155.47 us/run - 117.44 MFLOP/run - 755.39 GFLOPS
  28. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 15336 runs - 67.09 us/run - 117.44 MFLOP/run -  1.75 TFLOPS
  29. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11502 runs - 89.49 us/run - 234.88 MFLOP/run -  2.62 TFLOPS
  30. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11928 runs - 85.45 us/run - 234.88 MFLOP/run -  2.75 TFLOPS
  31. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8946 runs - 113.00 us/run - 234.88 MFLOP/run -  2.08 TFLOPS
  32. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8094 runs - 126.80 us/run - 234.88 MFLOP/run -  1.85 TFLOPS
  33. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8520 runs - 119.72 us/run - 234.88 MFLOP/run -  1.96 TFLOPS
  34. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2130 runs - 525.18 us/run - 234.88 MFLOP/run - 447.24 GFLOPS
  35. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5538 runs - 191.62 us/run - 234.88 MFLOP/run -  1.23 TFLOPS
  36. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 11076 runs - 93.37 us/run - 234.88 MFLOP/run -  2.52 TFLOPS
  37. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9798 runs - 103.32 us/run - 234.88 MFLOP/run -  2.27 TFLOPS
  38. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 10224 runs - 99.24 us/run - 234.88 MFLOP/run -  2.37 TFLOPS
  39. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8520 runs - 119.11 us/run - 234.88 MFLOP/run -  1.97 TFLOPS
  40. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7668 runs - 132.16 us/run - 234.88 MFLOP/run -  1.78 TFLOPS
  41. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7242 runs - 142.83 us/run - 234.88 MFLOP/run -  1.64 TFLOPS
  42. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8094 runs - 128.74 us/run - 234.88 MFLOP/run -  1.82 TFLOPS
  43. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7242 runs - 142.01 us/run - 234.88 MFLOP/run -  1.65 TFLOPS
  44. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6816 runs - 146.97 us/run - 234.88 MFLOP/run -  1.60 TFLOPS
  45. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6390 runs - 159.43 us/run - 234.88 MFLOP/run -  1.47 TFLOPS
  46. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6816 runs - 154.83 us/run - 234.88 MFLOP/run -  1.52 TFLOPS
  47. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8946 runs - 112.94 us/run - 234.88 MFLOP/run -  2.08 TFLOPS
  48. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 10650 runs - 95.43 us/run - 234.88 MFLOP/run -  2.46 TFLOPS
  49. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6816 runs - 155.57 us/run - 234.88 MFLOP/run -  1.51 TFLOPS
  50. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9372 runs - 109.36 us/run - 234.88 MFLOP/run -  2.15 TFLOPS
  51. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8804 runs - 117.17 us/run - 352.32 MFLOP/run -  3.01 TFLOPS
  52. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8236 runs - 122.64 us/run - 352.32 MFLOP/run -  2.87 TFLOPS
  53. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6532 runs - 153.74 us/run - 352.32 MFLOP/run -  2.29 TFLOPS
  54. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7384 runs - 136.55 us/run - 352.32 MFLOP/run -  2.58 TFLOPS
  55. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7100 runs - 143.95 us/run - 352.32 MFLOP/run -  2.45 TFLOPS
  56. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1420 runs - 876.09 us/run - 352.32 MFLOP/run - 402.15 GFLOPS
  57. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1988 runs - 532.33 us/run - 352.32 MFLOP/run - 661.84 GFLOPS
  58. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8804 runs - 117.35 us/run - 352.32 MFLOP/run -  3.00 TFLOPS
  59. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7668 runs - 132.19 us/run - 352.32 MFLOP/run -  2.67 TFLOPS
  60. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7100 runs - 142.49 us/run - 352.32 MFLOP/run -  2.47 TFLOPS
  61. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6532 runs - 156.14 us/run - 352.32 MFLOP/run -  2.26 TFLOPS
  62. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5396 runs - 192.52 us/run - 352.32 MFLOP/run -  1.83 TFLOPS
  63. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5112 runs - 202.53 us/run - 352.32 MFLOP/run -  1.74 TFLOPS
  64. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6532 runs - 158.21 us/run - 352.32 MFLOP/run -  2.23 TFLOPS
  65. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6532 runs - 155.57 us/run - 352.32 MFLOP/run -  2.26 TFLOPS
  66. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6248 runs - 162.21 us/run - 352.32 MFLOP/run -  2.17 TFLOPS
  67. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5680 runs - 176.12 us/run - 352.32 MFLOP/run -  2.00 TFLOPS
  68. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5964 runs - 169.23 us/run - 352.32 MFLOP/run -  2.08 TFLOPS
  69. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8804 runs - 114.81 us/run - 352.32 MFLOP/run -  3.07 TFLOPS
  70. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7952 runs - 126.69 us/run - 352.32 MFLOP/run -  2.78 TFLOPS
  71. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5680 runs - 182.05 us/run - 352.32 MFLOP/run -  1.94 TFLOPS
  72. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6532 runs - 154.55 us/run - 352.32 MFLOP/run -  2.28 TFLOPS
  73. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7242 runs - 140.93 us/run - 469.76 MFLOP/run -  3.33 TFLOPS
  74. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7668 runs - 130.79 us/run - 469.76 MFLOP/run -  3.59 TFLOPS
  75. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6177 runs - 163.04 us/run - 469.76 MFLOP/run -  2.88 TFLOPS
  76. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6390 runs - 161.87 us/run - 469.76 MFLOP/run -  2.90 TFLOPS
  77. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5751 runs - 176.35 us/run - 469.76 MFLOP/run -  2.66 TFLOPS
  78. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1065 runs - 989.29 us/run - 469.76 MFLOP/run - 474.85 GFLOPS
  79. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2556 runs - 403.86 us/run - 469.76 MFLOP/run -  1.16 TFLOPS
  80. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7242 runs - 139.91 us/run - 469.76 MFLOP/run -  3.36 TFLOPS
  81. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5538 runs - 185.31 us/run - 469.76 MFLOP/run -  2.53 TFLOPS
  82. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5538 runs - 182.33 us/run - 469.76 MFLOP/run -  2.58 TFLOPS
  83. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5751 runs - 179.05 us/run - 469.76 MFLOP/run -  2.62 TFLOPS
  84. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4260 runs - 235.32 us/run - 469.76 MFLOP/run -  2.00 TFLOPS
  85. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4260 runs - 245.95 us/run - 469.76 MFLOP/run -  1.91 TFLOPS
  86. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5751 runs - 180.30 us/run - 469.76 MFLOP/run -  2.61 TFLOPS
  87. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5964 runs - 170.44 us/run - 469.76 MFLOP/run -  2.76 TFLOPS
  88. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5751 runs - 178.56 us/run - 469.76 MFLOP/run -  2.63 TFLOPS
  89. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4899 runs - 211.82 us/run - 469.76 MFLOP/run -  2.22 TFLOPS
  90. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5538 runs - 183.97 us/run - 469.76 MFLOP/run -  2.55 TFLOPS
  91. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5112 runs - 201.00 us/run - 469.76 MFLOP/run -  2.34 TFLOPS
  92. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7029 runs - 144.24 us/run - 469.76 MFLOP/run -  3.26 TFLOPS
  93. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5112 runs - 197.19 us/run - 469.76 MFLOP/run -  2.38 TFLOPS
  94. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7668 runs - 132.81 us/run - 469.76 MFLOP/run -  3.54 TFLOPS
  95. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7011 runs - 144.93 us/run - 587.20 MFLOP/run -  4.05 TFLOPS
  96. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7182 runs - 140.26 us/run - 587.20 MFLOP/run -  4.19 TFLOPS
  97. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5814 runs - 175.82 us/run - 587.20 MFLOP/run -  3.34 TFLOPS
  98. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5985 runs - 169.47 us/run - 587.20 MFLOP/run -  3.46 TFLOPS
  99. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5472 runs - 188.02 us/run - 587.20 MFLOP/run -  3.12 TFLOPS
  100. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1026 runs - 1038.50 us/run - 587.20 MFLOP/run - 565.43 GFLOPS
  101. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2907 runs - 361.76 us/run - 587.20 MFLOP/run -  1.62 TFLOPS
  102. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6840 runs - 146.38 us/run - 587.20 MFLOP/run -  4.01 TFLOPS
  103. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5472 runs - 187.58 us/run - 587.20 MFLOP/run -  3.13 TFLOPS
  104. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4959 runs - 202.35 us/run - 587.20 MFLOP/run -  2.90 TFLOPS
  105. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5130 runs - 196.35 us/run - 587.20 MFLOP/run -  2.99 TFLOPS
  106. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3249 runs - 324.49 us/run - 587.20 MFLOP/run -  1.81 TFLOPS
  107. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3078 runs - 332.91 us/run - 587.20 MFLOP/run -  1.76 TFLOPS
  108. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5130 runs - 196.67 us/run - 587.20 MFLOP/run -  2.99 TFLOPS
  109. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5643 runs - 177.78 us/run - 587.20 MFLOP/run -  3.30 TFLOPS
  110. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5301 runs - 189.29 us/run - 587.20 MFLOP/run -  3.10 TFLOPS
  111. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4617 runs - 219.19 us/run - 587.20 MFLOP/run -  2.68 TFLOPS
  112. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5301 runs - 193.96 us/run - 587.20 MFLOP/run -  3.03 TFLOPS
  113. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 6669 runs - 151.51 us/run - 587.20 MFLOP/run -  3.88 TFLOPS
  114. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7011 runs - 145.34 us/run - 587.20 MFLOP/run -  4.04 TFLOPS
  115. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5130 runs - 200.82 us/run - 587.20 MFLOP/run -  2.92 TFLOPS
  116. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7524 runs - 135.87 us/run - 587.20 MFLOP/run -  4.32 TFLOPS
  117. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5350 runs - 187.24 us/run - 939.52 MFLOP/run -  5.02 TFLOPS
  118. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5564 runs - 181.11 us/run - 939.52 MFLOP/run -  5.19 TFLOPS
  119. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4815 runs - 209.35 us/run - 939.52 MFLOP/run -  4.49 TFLOPS
  120. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5136 runs - 196.89 us/run - 939.52 MFLOP/run -  4.77 TFLOPS
  121. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3210 runs - 312.10 us/run - 939.52 MFLOP/run -  3.01 TFLOPS
  122. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 642 runs - 1677.33 us/run - 939.52 MFLOP/run - 560.13 GFLOPS
  123. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1926 runs - 526.66 us/run - 939.52 MFLOP/run -  1.78 TFLOPS
  124. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5243 runs - 194.70 us/run - 939.52 MFLOP/run -  4.83 TFLOPS
  125. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2568 runs - 401.12 us/run - 939.52 MFLOP/run -  2.34 TFLOPS
  126. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3210 runs - 316.31 us/run - 939.52 MFLOP/run -  2.97 TFLOPS
  127. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3531 runs - 289.44 us/run - 939.52 MFLOP/run -  3.25 TFLOPS
  128. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2140 runs - 483.61 us/run - 939.52 MFLOP/run -  1.94 TFLOPS
  129. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1177 runs - 883.75 us/run - 939.52 MFLOP/run -  1.06 TFLOPS
  130. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2889 runs - 350.10 us/run - 939.52 MFLOP/run -  2.68 TFLOPS
  131. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4494 runs - 224.34 us/run - 939.52 MFLOP/run -  4.19 TFLOPS
  132. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4066 runs - 251.89 us/run - 939.52 MFLOP/run -  3.73 TFLOPS
  133. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3638 runs - 278.32 us/run - 939.52 MFLOP/run -  3.38 TFLOPS
  134. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3959 runs - 254.27 us/run - 939.52 MFLOP/run -  3.69 TFLOPS
  135. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4922 runs - 206.66 us/run - 939.52 MFLOP/run -  4.55 TFLOPS
  136. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5243 runs - 193.79 us/run - 939.52 MFLOP/run -  4.85 TFLOPS
  137. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 4173 runs - 244.21 us/run - 939.52 MFLOP/run -  3.85 TFLOPS
  138. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5029 runs - 202.42 us/run - 939.52 MFLOP/run -  4.64 TFLOPS
  139. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2970 runs - 337.73 us/run - 1.88 GFLOP/run -  5.56 TFLOPS
  140. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3132 runs - 324.77 us/run - 1.88 GFLOP/run -  5.79 TFLOPS
  141. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1782 runs - 575.96 us/run - 1.88 GFLOP/run -  3.26 TFLOPS
  142. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2106 runs - 485.83 us/run - 1.88 GFLOP/run -  3.87 TFLOPS
  143. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1026 runs - 1001.61 us/run - 1.88 GFLOP/run -  1.88 TFLOPS
  144. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 810 runs - 1318.23 us/run - 1.88 GFLOP/run -  1.43 TFLOPS
  145. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 756 runs - 1424.23 us/run - 1.88 GFLOP/run -  1.32 TFLOPS
  146. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2322 runs - 438.83 us/run - 1.88 GFLOP/run -  4.28 TFLOPS
  147. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 621.84 us/run - 1.88 GFLOP/run -  3.02 TFLOPS
  148. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2376 runs - 430.23 us/run - 1.88 GFLOP/run -  4.37 TFLOPS
  149. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2160 runs - 472.81 us/run - 1.88 GFLOP/run -  3.97 TFLOPS
  150. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2484 runs - 407.15 us/run - 1.88 GFLOP/run -  4.62 TFLOPS
  151. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2430 runs - 414.17 us/run - 1.88 GFLOP/run -  4.54 TFLOPS
  152. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2214 runs - 453.42 us/run - 1.88 GFLOP/run -  4.14 TFLOPS
  153. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1728 runs - 582.25 us/run - 1.88 GFLOP/run -  3.23 TFLOPS
  154. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1566 runs - 646.01 us/run - 1.88 GFLOP/run -  2.91 TFLOPS
  155. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1674 runs - 609.69 us/run - 1.88 GFLOP/run -  3.08 TFLOPS
  156. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1566 runs - 660.21 us/run - 1.88 GFLOP/run -  2.85 TFLOPS
  157. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2322 runs - 437.92 us/run - 1.88 GFLOP/run -  4.29 TFLOPS
  158. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2322 runs - 438.27 us/run - 1.88 GFLOP/run -  4.29 TFLOPS
  159. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1674 runs - 597.64 us/run - 1.88 GFLOP/run -  3.14 TFLOPS
  160. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2376 runs - 428.04 us/run - 1.88 GFLOP/run -  4.39 TFLOPS
  161. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2268 runs - 443.38 us/run - 2.82 GFLOP/run -  6.36 TFLOPS
  162. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2412 runs - 417.89 us/run - 2.82 GFLOP/run -  6.74 TFLOPS
  163. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1656 runs - 614.62 us/run - 2.82 GFLOP/run -  4.59 TFLOPS
  164. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1836 runs - 554.71 us/run - 2.82 GFLOP/run -  5.08 TFLOPS
  165. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2052 runs - 490.58 us/run - 2.82 GFLOP/run -  5.75 TFLOPS
  166. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1152 runs - 883.04 us/run - 2.82 GFLOP/run -  3.19 TFLOPS
  167. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1152 runs - 892.09 us/run - 2.82 GFLOP/run -  3.16 TFLOPS
  168. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2016 runs - 502.30 us/run - 2.82 GFLOP/run -  5.61 TFLOPS
  169. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1728 runs - 580.95 us/run - 2.82 GFLOP/run -  4.85 TFLOPS
  170. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1440 runs - 695.98 us/run - 2.82 GFLOP/run -  4.05 TFLOPS
  171. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 624.90 us/run - 2.82 GFLOP/run -  4.51 TFLOPS
  172. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1872 runs - 539.09 us/run - 2.82 GFLOP/run -  5.23 TFLOPS
  173. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1800 runs - 557.15 us/run - 2.82 GFLOP/run -  5.06 TFLOPS
  174. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 620.44 us/run - 2.82 GFLOP/run -  4.54 TFLOPS
  175. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 622.77 us/run - 2.82 GFLOP/run -  4.53 TFLOPS
  176. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1548 runs - 648.30 us/run - 2.82 GFLOP/run -  4.35 TFLOPS
  177. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 671.00 us/run - 2.82 GFLOP/run -  4.20 TFLOPS
  178. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 669.64 us/run - 2.82 GFLOP/run -  4.21 TFLOPS
  179. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1908 runs - 524.25 us/run - 2.82 GFLOP/run -  5.38 TFLOPS
  180. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2016 runs - 500.52 us/run - 2.82 GFLOP/run -  5.63 TFLOPS
  181. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1476 runs - 690.54 us/run - 2.82 GFLOP/run -  4.08 TFLOPS
  182. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1980 runs - 506.75 us/run - 2.82 GFLOP/run -  5.56 TFLOPS
  183. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1944 runs - 517.45 us/run - 3.76 GFLOP/run -  7.26 TFLOPS
  184. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1998 runs - 502.93 us/run - 3.76 GFLOP/run -  7.47 TFLOPS
  185. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1377 runs - 730.67 us/run - 3.76 GFLOP/run -  5.14 TFLOPS
  186. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1539 runs - 655.04 us/run - 3.76 GFLOP/run -  5.74 TFLOPS
  187. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1782 runs - 566.59 us/run - 3.76 GFLOP/run -  6.63 TFLOPS
  188. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 999 runs - 1024.56 us/run - 3.76 GFLOP/run -  3.67 TFLOPS
  189. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1026 runs - 986.72 us/run - 3.76 GFLOP/run -  3.81 TFLOPS
  190. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 623.44 us/run - 3.76 GFLOP/run -  6.03 TFLOPS
  191. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 787.03 us/run - 3.76 GFLOP/run -  4.78 TFLOPS
  192. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 999 runs - 1006.06 us/run - 3.76 GFLOP/run -  3.74 TFLOPS
  193. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1404 runs - 720.73 us/run - 3.76 GFLOP/run -  5.21 TFLOPS
  194. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1593 runs - 633.93 us/run - 3.76 GFLOP/run -  5.93 TFLOPS
  195. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1404 runs - 719.52 us/run - 3.76 GFLOP/run -  5.22 TFLOPS
  196. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 771.67 us/run - 3.76 GFLOP/run -  4.87 TFLOPS
  197. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1323 runs - 763.69 us/run - 3.76 GFLOP/run -  4.92 TFLOPS
  198. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 778.89 us/run - 3.76 GFLOP/run -  4.82 TFLOPS
  199. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 793.00 us/run - 3.76 GFLOP/run -  4.74 TFLOPS
  200. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 791.01 us/run - 3.76 GFLOP/run -  4.75 TFLOPS
  201. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1539 runs - 650.90 us/run - 3.76 GFLOP/run -  5.77 TFLOPS
  202. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 623.61 us/run - 3.76 GFLOP/run -  6.03 TFLOPS
  203. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1242 runs - 815.20 us/run - 3.76 GFLOP/run -  4.61 TFLOPS
  204. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 626.21 us/run - 3.76 GFLOP/run -  6.00 TFLOPS
  205. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1672 runs - 602.98 us/run - 4.70 GFLOP/run -  7.79 TFLOPS
  206. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1738 runs - 581.11 us/run - 4.70 GFLOP/run -  8.08 TFLOPS
  207. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1188 runs - 848.60 us/run - 4.70 GFLOP/run -  5.54 TFLOPS
  208. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1298 runs - 777.47 us/run - 4.70 GFLOP/run -  6.04 TFLOPS
  209. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1540 runs - 657.21 us/run - 4.70 GFLOP/run -  7.15 TFLOPS
  210. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 858 runs - 1174.94 us/run - 4.70 GFLOP/run -  4.00 TFLOPS
  211. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 924 runs - 1085.28 us/run - 4.70 GFLOP/run -  4.33 TFLOPS
  212. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1320 runs - 760.74 us/run - 4.70 GFLOP/run -  6.18 TFLOPS
  213. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1056 runs - 949.81 us/run - 4.70 GFLOP/run -  4.95 TFLOPS
  214. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 990 runs - 1018.65 us/run - 4.70 GFLOP/run -  4.61 TFLOPS
  215. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1144 runs - 876.26 us/run - 4.70 GFLOP/run -  5.36 TFLOPS
  216. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1408 runs - 721.43 us/run - 4.70 GFLOP/run -  6.51 TFLOPS
  217. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1320 runs - 767.44 us/run - 4.70 GFLOP/run -  6.12 TFLOPS
  218. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1342 runs - 751.29 us/run - 4.70 GFLOP/run -  6.25 TFLOPS
  219. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1144 runs - 884.73 us/run - 4.70 GFLOP/run -  5.31 TFLOPS
  220. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1122 runs - 900.32 us/run - 4.70 GFLOP/run -  5.22 TFLOPS
  221. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1100 runs - 914.31 us/run - 4.70 GFLOP/run -  5.14 TFLOPS
  222. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1100 runs - 909.28 us/run - 4.70 GFLOP/run -  5.17 TFLOPS
  223. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1188 runs - 850.17 us/run - 4.70 GFLOP/run -  5.53 TFLOPS
  224. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1342 runs - 745.54 us/run - 4.70 GFLOP/run -  6.30 TFLOPS
  225. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1100 runs - 924.19 us/run - 4.70 GFLOP/run -  5.08 TFLOPS
  226. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1320 runs - 763.99 us/run - 4.70 GFLOP/run -  6.15 TFLOPS
  227. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1458 runs - 687.80 us/run - 5.64 GFLOP/run -  8.20 TFLOPS
  228. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 665.85 us/run - 5.64 GFLOP/run -  8.47 TFLOPS
  229. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1026 runs - 992.01 us/run - 5.64 GFLOP/run -  5.68 TFLOPS
  230. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1116 runs - 898.74 us/run - 5.64 GFLOP/run -  6.27 TFLOPS
  231. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1350 runs - 746.89 us/run - 5.64 GFLOP/run -  7.55 TFLOPS
  232. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 774 runs - 1302.52 us/run - 5.64 GFLOP/run -  4.33 TFLOPS
  233. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 846 runs - 1188.08 us/run - 5.64 GFLOP/run -  4.74 TFLOPS
  234. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1170 runs - 867.36 us/run - 5.64 GFLOP/run -  6.50 TFLOPS
  235. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1062 runs - 951.09 us/run - 5.64 GFLOP/run -  5.93 TFLOPS
  236. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 396 runs - 2533.64 us/run - 5.64 GFLOP/run -  2.22 TFLOPS
  237. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 954 runs - 1049.46 us/run - 5.64 GFLOP/run -  5.37 TFLOPS
  238. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1170 runs - 860.39 us/run - 5.64 GFLOP/run -  6.55 TFLOPS
  239. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 936 runs - 1083.83 us/run - 5.64 GFLOP/run -  5.20 TFLOPS
  240. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1152 runs - 869.33 us/run - 5.64 GFLOP/run -  6.48 TFLOPS
  241. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1008 runs - 1007.87 us/run - 5.64 GFLOP/run -  5.59 TFLOPS
  242. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 990 runs - 1024.41 us/run - 5.64 GFLOP/run -  5.50 TFLOPS
  243. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 972 runs - 1034.25 us/run - 5.64 GFLOP/run -  5.45 TFLOPS
  244. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 990 runs - 1020.37 us/run - 5.64 GFLOP/run -  5.52 TFLOPS
  245. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1170 runs - 857.96 us/run - 5.64 GFLOP/run -  6.57 TFLOPS
  246. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1152 runs - 871.54 us/run - 5.64 GFLOP/run -  6.47 TFLOPS
  247. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 972 runs - 1038.49 us/run - 5.64 GFLOP/run -  5.43 TFLOPS
  248. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1170 runs - 863.13 us/run - 5.64 GFLOP/run -  6.53 TFLOPS
  249. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1148 runs - 875.45 us/run - 7.52 GFLOP/run -  8.59 TFLOPS
  250. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1204 runs - 837.07 us/run - 7.52 GFLOP/run -  8.98 TFLOPS
  251. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 840 runs - 1208.01 us/run - 7.52 GFLOP/run -  6.22 TFLOPS
  252. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 868 runs - 1162.22 us/run - 7.52 GFLOP/run -  6.47 TFLOPS
  253. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1064 runs - 940.42 us/run - 7.52 GFLOP/run -  7.99 TFLOPS
  254. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 588 runs - 1724.21 us/run - 7.52 GFLOP/run -  4.36 TFLOPS
  255. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 714 runs - 1408.90 us/run - 7.52 GFLOP/run -  5.33 TFLOPS
  256. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 910 runs - 1113.32 us/run - 7.52 GFLOP/run -  6.75 TFLOPS
  257. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 854 runs - 1183.76 us/run - 7.52 GFLOP/run -  6.35 TFLOPS
  258. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 210 runs - 5096.61 us/run - 7.52 GFLOP/run -  1.47 TFLOPS
  259. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 714 runs - 1427.01 us/run - 7.52 GFLOP/run -  5.27 TFLOPS
  260. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 686 runs - 1464.78 us/run - 7.52 GFLOP/run -  5.13 TFLOPS
  261. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 644 runs - 1562.53 us/run - 7.52 GFLOP/run -  4.81 TFLOPS
  262. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 910 runs - 1103.13 us/run - 7.52 GFLOP/run -  6.81 TFLOPS
  263. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 826 runs - 1224.80 us/run - 7.52 GFLOP/run -  6.14 TFLOPS
  264. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 798 runs - 1260.20 us/run - 7.52 GFLOP/run -  5.96 TFLOPS
  265. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 784 runs - 1279.43 us/run - 7.52 GFLOP/run -  5.87 TFLOPS
  266. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 812 runs - 1249.03 us/run - 7.52 GFLOP/run -  6.02 TFLOPS
  267. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 896 runs - 1121.18 us/run - 7.52 GFLOP/run -  6.70 TFLOPS
  268. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 896 runs - 1116.71 us/run - 7.52 GFLOP/run -  6.73 TFLOPS
  269. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 784 runs - 1277.34 us/run - 7.52 GFLOP/run -  5.88 TFLOPS
  270. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 910 runs - 1099.50 us/run - 7.52 GFLOP/run -  6.84 TFLOPS
  271. MUL_MAT(type_a=q4_0,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 272 runs - 3693.92 us/run - 60.13 GFLOP/run -  16.28 TFLOPS
  272. MUL_MAT(type_a=q4_1,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 270 runs - 3704.30 us/run - 60.13 GFLOP/run -  16.23 TFLOPS
  273. MUL_MAT(type_a=q5_0,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 170 runs - 5930.80 us/run - 60.13 GFLOP/run -  10.14 TFLOPS
  274. MUL_MAT(type_a=q5_1,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 178 runs - 5681.16 us/run - 60.13 GFLOP/run -  10.58 TFLOPS
  275. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 186 runs - 5429.75 us/run - 60.13 GFLOP/run -  11.07 TFLOPS
  276. MUL_MAT(type_a=q1_0,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 98 runs - 10408.05 us/run - 60.13 GFLOP/run -  5.78 TFLOPS
  277. MUL_MAT(type_a=q2_0,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 138 runs - 7316.62 us/run - 60.13 GFLOP/run -  8.22 TFLOPS
  278. MUL_MAT(type_a=mxfp4,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 180 runs - 5613.04 us/run - 60.13 GFLOP/run -  10.71 TFLOPS
  279. MUL_MAT(type_a=nvfp4,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 168 runs - 5993.55 us/run - 60.13 GFLOP/run -  10.03 TFLOPS
  280. MUL_MAT(type_a=q2_K,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 34 runs - 30139.35 us/run - 60.13 GFLOP/run -  2.00 TFLOPS
  281. MUL_MAT(type_a=q3_K,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 146 runs - 6898.00 us/run - 60.13 GFLOP/run -  8.72 TFLOPS
  282. MUL_MAT(type_a=q4_K,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 198 runs - 5074.96 us/run - 60.13 GFLOP/run -  11.85 TFLOPS
  283. MUL_MAT(type_a=q5_K,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 126 runs - 7942.40 us/run - 60.13 GFLOP/run -  7.57 TFLOPS
  284. MUL_MAT(type_a=q6_K,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 190 runs - 5306.27 us/run - 60.13 GFLOP/run -  11.33 TFLOPS
  285. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 170 runs - 5910.80 us/run - 60.13 GFLOP/run -  10.17 TFLOPS
  286. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 164 runs - 6166.38 us/run - 60.13 GFLOP/run -  9.75 TFLOPS
  287. MUL_MAT(type_a=iq2_s,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 162 runs - 6225.80 us/run - 60.13 GFLOP/run -  9.66 TFLOPS
  288. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 168 runs - 6024.48 us/run - 60.13 GFLOP/run -  9.98 TFLOPS
  289. MUL_MAT(type_a=iq1_s,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 182 runs - 5532.08 us/run - 60.13 GFLOP/run -  10.87 TFLOPS
  290. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 180 runs - 5600.58 us/run - 60.13 GFLOP/run -  10.74 TFLOPS
  291. MUL_MAT(type_a=iq3_s,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 164 runs - 6148.37 us/run - 60.13 GFLOP/run -  9.78 TFLOPS
  292. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 180 runs - 5559.53 us/run - 60.13 GFLOP/run -  10.82 TFLOPS
  293. MUL_MAT(type_a=q4_0,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2970 runs - 339.27 us/run - 1.88 GFLOP/run -  5.54 TFLOPS
  294. MUL_MAT(type_a=q4_1,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3132 runs - 320.64 us/run - 1.88 GFLOP/run -  5.86 TFLOPS
  295. MUL_MAT(type_a=q5_0,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1998 runs - 501.18 us/run - 1.88 GFLOP/run -  3.75 TFLOPS
  296. MUL_MAT(type_a=q5_1,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2430 runs - 412.33 us/run - 1.88 GFLOP/run -  4.56 TFLOPS
  297. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2322 runs - 440.81 us/run - 1.88 GFLOP/run -  4.26 TFLOPS
  298. MUL_MAT(type_a=q1_0,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 702 runs - 1448.73 us/run - 1.88 GFLOP/run -  1.30 TFLOPS
  299. MUL_MAT(type_a=q2_0,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 702 runs - 1426.02 us/run - 1.88 GFLOP/run -  1.32 TFLOPS
  300. MUL_MAT(type_a=mxfp4,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2754 runs - 365.44 us/run - 1.88 GFLOP/run -  5.14 TFLOPS
  301. MUL_MAT(type_a=nvfp4,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1674 runs - 603.48 us/run - 1.88 GFLOP/run -  3.11 TFLOPS
  302. MUL_MAT(type_a=q2_K,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2268 runs - 445.49 us/run - 1.88 GFLOP/run -  4.22 TFLOPS
  303. MUL_MAT(type_a=q3_K,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2268 runs - 441.98 us/run - 1.88 GFLOP/run -  4.25 TFLOPS
  304. MUL_MAT(type_a=q4_K,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2538 runs - 399.54 us/run - 1.88 GFLOP/run -  4.70 TFLOPS
  305. MUL_MAT(type_a=q5_K,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2538 runs - 401.79 us/run - 1.88 GFLOP/run -  4.68 TFLOPS
  306. MUL_MAT(type_a=q6_K,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2430 runs - 417.04 us/run - 1.88 GFLOP/run -  4.51 TFLOPS
  307. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1836 runs - 555.46 us/run - 1.88 GFLOP/run -  3.38 TFLOPS
  308. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 631.09 us/run - 1.88 GFLOP/run -  2.98 TFLOPS
  309. MUL_MAT(type_a=iq2_s,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1242 runs - 814.99 us/run - 1.88 GFLOP/run -  2.31 TFLOPS
  310. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 623.92 us/run - 1.88 GFLOP/run -  3.01 TFLOPS
  311. MUL_MAT(type_a=iq1_s,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2376 runs - 421.08 us/run - 1.88 GFLOP/run -  4.46 TFLOPS
  312. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2808 runs - 361.88 us/run - 1.88 GFLOP/run -  5.19 TFLOPS
  313. MUL_MAT(type_a=iq3_s,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1728 runs - 589.85 us/run - 1.88 GFLOP/run -  3.19 TFLOPS
  314. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1404 runs - 723.29 us/run - 1.88 GFLOP/run -  2.60 TFLOPS
  315. MUL_MAT(type_a=q4_0,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1971 runs - 511.86 us/run - 3.76 GFLOP/run -  7.34 TFLOPS
  316. MUL_MAT(type_a=q4_1,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2025 runs - 499.80 us/run - 3.76 GFLOP/run -  7.52 TFLOPS
  317. MUL_MAT(type_a=q5_0,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1377 runs - 733.45 us/run - 3.76 GFLOP/run -  5.13 TFLOPS
  318. MUL_MAT(type_a=q5_1,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1458 runs - 698.36 us/run - 3.76 GFLOP/run -  5.38 TFLOPS
  319. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 665.44 us/run - 3.76 GFLOP/run -  5.65 TFLOPS
  320. MUL_MAT(type_a=q1_0,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 702 runs - 1439.60 us/run - 3.76 GFLOP/run -  2.61 TFLOPS
  321. MUL_MAT(type_a=q2_0,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 945 runs - 1059.61 us/run - 3.76 GFLOP/run -  3.55 TFLOPS
  322. MUL_MAT(type_a=mxfp4,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 622.44 us/run - 3.76 GFLOP/run -  6.04 TFLOPS
  323. MUL_MAT(type_a=nvfp4,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 793.07 us/run - 3.76 GFLOP/run -  4.74 TFLOPS
  324. MUL_MAT(type_a=q2_K,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1080 runs - 940.81 us/run - 3.76 GFLOP/run -  4.00 TFLOPS
  325. MUL_MAT(type_a=q3_K,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1404 runs - 723.40 us/run - 3.76 GFLOP/run -  5.20 TFLOPS
  326. MUL_MAT(type_a=q4_K,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1485 runs - 685.76 us/run - 3.76 GFLOP/run -  5.48 TFLOPS
  327. MUL_MAT(type_a=q5_K,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1431 runs - 710.91 us/run - 3.76 GFLOP/run -  5.29 TFLOPS
  328. MUL_MAT(type_a=q6_K,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1593 runs - 631.44 us/run - 3.76 GFLOP/run -  5.95 TFLOPS
  329. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1323 runs - 760.76 us/run - 3.76 GFLOP/run -  4.94 TFLOPS
  330. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 781.59 us/run - 3.76 GFLOP/run -  4.81 TFLOPS
  331. MUL_MAT(type_a=iq2_s,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 797.16 us/run - 3.76 GFLOP/run -  4.72 TFLOPS
  332. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 778.29 us/run - 3.76 GFLOP/run -  4.83 TFLOPS
  333. MUL_MAT(type_a=iq1_s,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1566 runs - 646.53 us/run - 3.76 GFLOP/run -  5.81 TFLOPS
  334. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 619.01 us/run - 3.76 GFLOP/run -  6.07 TFLOPS
  335. MUL_MAT(type_a=iq3_s,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 804.64 us/run - 3.76 GFLOP/run -  4.67 TFLOPS
  336. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 623.27 us/run - 3.76 GFLOP/run -  6.03 TFLOPS
  337. MUL_MAT(type_a=q4_0,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1148 runs - 875.74 us/run - 7.52 GFLOP/run -  8.58 TFLOPS
  338. MUL_MAT(type_a=q4_1,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1204 runs - 834.97 us/run - 7.52 GFLOP/run -  9.00 TFLOPS
  339. MUL_MAT(type_a=q5_0,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 840 runs - 1201.26 us/run - 7.52 GFLOP/run -  6.26 TFLOPS
  340. MUL_MAT(type_a=q5_1,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 896 runs - 1131.38 us/run - 7.52 GFLOP/run -  6.65 TFLOPS
  341. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 910 runs - 1100.32 us/run - 7.52 GFLOP/run -  6.83 TFLOPS
  342. MUL_MAT(type_a=q1_0,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 644 runs - 1583.35 us/run - 7.52 GFLOP/run -  4.75 TFLOPS
  343. MUL_MAT(type_a=q2_0,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 700 runs - 1444.37 us/run - 7.52 GFLOP/run -  5.21 TFLOPS
  344. MUL_MAT(type_a=mxfp4,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 938 runs - 1068.00 us/run - 7.52 GFLOP/run -  7.04 TFLOPS
  345. MUL_MAT(type_a=nvfp4,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 854 runs - 1181.73 us/run - 7.52 GFLOP/run -  6.36 TFLOPS
  346. MUL_MAT(type_a=q2_K,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 196 runs - 5276.01 us/run - 7.52 GFLOP/run -  1.42 TFLOPS
  347. MUL_MAT(type_a=q3_K,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 770 runs - 1312.66 us/run - 7.52 GFLOP/run -  5.73 TFLOPS
  348. MUL_MAT(type_a=q4_K,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 854 runs - 1173.65 us/run - 7.52 GFLOP/run -  6.41 TFLOPS
  349. MUL_MAT(type_a=q5_K,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 980 runs - 1033.52 us/run - 7.52 GFLOP/run -  7.27 TFLOPS
  350. MUL_MAT(type_a=q6_K,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 798 runs - 1269.75 us/run - 7.52 GFLOP/run -  5.92 TFLOPS
  351. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 826 runs - 1221.59 us/run - 7.52 GFLOP/run -  6.15 TFLOPS
  352. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 798 runs - 1256.02 us/run - 7.52 GFLOP/run -  5.99 TFLOPS
  353. MUL_MAT(type_a=iq2_s,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 784 runs - 1283.56 us/run - 7.52 GFLOP/run -  5.86 TFLOPS
  354. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 812 runs - 1245.58 us/run - 7.52 GFLOP/run -  6.04 TFLOPS
  355. MUL_MAT(type_a=iq1_s,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 910 runs - 1112.31 us/run - 7.52 GFLOP/run -  6.76 TFLOPS
  356. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 952 runs - 1065.05 us/run - 7.52 GFLOP/run -  7.06 TFLOPS
  357. MUL_MAT(type_a=iq3_s,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 826 runs - 1228.29 us/run - 7.52 GFLOP/run -  6.12 TFLOPS
  358. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 952 runs - 1063.55 us/run - 7.52 GFLOP/run -  7.07 TFLOPS
  359. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=129,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 273 runs - 3726.88 us/run - 15.15 GFLOP/run -  4.07 TFLOPS
  360. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=145,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 438 runs - 2288.45 us/run - 17.03 GFLOP/run -  7.44 TFLOPS
  361. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=161,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 450 runs - 2246.66 us/run - 18.91 GFLOP/run -  8.42 TFLOPS
  362. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=177,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 345 runs - 2901.45 us/run - 20.79 GFLOP/run -  7.16 TFLOPS
  363. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=193,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 310 runs - 3275.13 us/run - 22.67 GFLOP/run -  6.92 TFLOPS
  364. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=209,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 285 runs - 3534.33 us/run - 24.55 GFLOP/run -  6.94 TFLOPS
  365. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=225,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 296 runs - 3401.97 us/run - 26.42 GFLOP/run -  7.77 TFLOPS
  366. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=241,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 284 runs - 3556.93 us/run - 28.30 GFLOP/run -  7.96 TFLOPS
  367. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=129,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 420 runs - 2416.19 us/run - 15.39 GFLOP/run -  6.37 TFLOPS
  368. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=145,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 546 runs - 1838.37 us/run - 17.29 GFLOP/run -  9.41 TFLOPS
  369. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=161,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 438 runs - 2298.38 us/run - 19.20 GFLOP/run -  8.36 TFLOPS
  370. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=177,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 335 runs - 3017.09 us/run - 21.11 GFLOP/run -  7.00 TFLOPS
  371. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=193,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 345 runs - 2935.02 us/run - 23.02 GFLOP/run -  7.84 TFLOPS
  372. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=209,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 330 runs - 3068.55 us/run - 24.93 GFLOP/run -  8.12 TFLOPS
  373. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=225,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 312 runs - 3211.85 us/run - 26.84 GFLOP/run -  8.36 TFLOPS
  374. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=241,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 300 runs - 3370.47 us/run - 28.75 GFLOP/run -  8.53 TFLOPS
  375. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=129,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 413 runs - 2441.50 us/run - 15.15 GFLOP/run -  6.21 TFLOPS
  376. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=145,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 546 runs - 1835.34 us/run - 17.03 GFLOP/run -  9.28 TFLOPS
  377. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=161,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 438 runs - 2299.85 us/run - 18.91 GFLOP/run -  8.22 TFLOPS
  378. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=177,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 330 runs - 3060.90 us/run - 20.79 GFLOP/run -  6.79 TFLOPS
  379. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=193,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 335 runs - 2990.04 us/run - 22.67 GFLOP/run -  7.58 TFLOPS
  380. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=209,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 325 runs - 3123.64 us/run - 24.55 GFLOP/run -  7.86 TFLOPS
  381. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=225,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 308 runs - 3279.02 us/run - 26.43 GFLOP/run -  8.06 TFLOPS
  382. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=241,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 292 runs - 3449.37 us/run - 28.31 GFLOP/run -  8.21 TFLOPS
  383. MUL_MAT(type_a=q4_0,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1944 runs - 516.83 us/run - 3.82 GFLOP/run -  7.39 TFLOPS
  384. MUL_MAT(type_a=q4_1,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1998 runs - 502.08 us/run - 3.82 GFLOP/run -  7.60 TFLOPS
  385. MUL_MAT(type_a=q5_0,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1377 runs - 730.31 us/run - 3.82 GFLOP/run -  5.23 TFLOPS
  386. MUL_MAT(type_a=q5_1,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1539 runs - 655.25 us/run - 3.82 GFLOP/run -  5.82 TFLOPS
  387. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 670.10 us/run - 3.82 GFLOP/run -  5.70 TFLOPS
  388. MUL_MAT(type_a=q1_0,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 702 runs - 1458.54 us/run - 3.82 GFLOP/run -  2.62 TFLOPS
  389. MUL_MAT(type_a=q2_0,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 945 runs - 1061.95 us/run - 3.82 GFLOP/run -  3.59 TFLOPS
  390. MUL_MAT(type_a=mxfp4,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 624.01 us/run - 3.82 GFLOP/run -  6.12 TFLOPS
  391. MUL_MAT(type_a=nvfp4,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 785.02 us/run - 3.82 GFLOP/run -  4.86 TFLOPS
  392. MUL_MAT(type_a=q2_K,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 999 runs - 1003.05 us/run - 3.82 GFLOP/run -  3.81 TFLOPS
  393. MUL_MAT(type_a=q3_K,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1404 runs - 723.34 us/run - 3.82 GFLOP/run -  5.28 TFLOPS
  394. MUL_MAT(type_a=q4_K,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1458 runs - 688.81 us/run - 3.82 GFLOP/run -  5.54 TFLOPS
  395. MUL_MAT(type_a=q5_K,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1404 runs - 718.76 us/run - 3.82 GFLOP/run -  5.31 TFLOPS
  396. MUL_MAT(type_a=q6_K,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 772.50 us/run - 3.82 GFLOP/run -  4.94 TFLOPS
  397. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1323 runs - 763.32 us/run - 3.82 GFLOP/run -  5.00 TFLOPS
  398. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1296 runs - 778.69 us/run - 3.82 GFLOP/run -  4.90 TFLOPS
  399. MUL_MAT(type_a=iq2_s,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 792.10 us/run - 3.82 GFLOP/run -  4.82 TFLOPS
  400. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1269 runs - 790.82 us/run - 3.82 GFLOP/run -  4.83 TFLOPS
  401. MUL_MAT(type_a=iq1_s,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1539 runs - 650.83 us/run - 3.82 GFLOP/run -  5.86 TFLOPS
  402. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 623.58 us/run - 3.82 GFLOP/run -  6.12 TFLOPS
  403. MUL_MAT(type_a=iq3_s,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1242 runs - 814.16 us/run - 3.82 GFLOP/run -  4.69 TFLOPS
  404. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1620 runs - 626.18 us/run - 3.82 GFLOP/run -  6.10 TFLOPS
  405. MUL_MAT(type_a=q4_0,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1148 runs - 876.84 us/run - 7.63 GFLOP/run -  8.71 TFLOPS
  406. MUL_MAT(type_a=q4_1,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1204 runs - 836.50 us/run - 7.63 GFLOP/run -  9.13 TFLOPS
  407. MUL_MAT(type_a=q5_0,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 840 runs - 1201.04 us/run - 7.63 GFLOP/run -  6.36 TFLOPS
  408. MUL_MAT(type_a=q5_1,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 896 runs - 1133.24 us/run - 7.63 GFLOP/run -  6.74 TFLOPS
  409. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 938 runs - 1081.44 us/run - 7.63 GFLOP/run -  7.06 TFLOPS
  410. MUL_MAT(type_a=q1_0,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 546 runs - 1854.04 us/run - 7.63 GFLOP/run -  4.12 TFLOPS
  411. MUL_MAT(type_a=q2_0,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 686 runs - 1462.08 us/run - 7.63 GFLOP/run -  5.22 TFLOPS
  412. MUL_MAT(type_a=mxfp4,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 938 runs - 1075.14 us/run - 7.63 GFLOP/run -  7.10 TFLOPS
  413. MUL_MAT(type_a=nvfp4,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 854 runs - 1184.19 us/run - 7.63 GFLOP/run -  6.45 TFLOPS
  414. MUL_MAT(type_a=q2_K,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 210 runs - 5086.60 us/run - 7.63 GFLOP/run -  1.50 TFLOPS
  415. MUL_MAT(type_a=q3_K,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 714 runs - 1425.80 us/run - 7.63 GFLOP/run -  5.35 TFLOPS
  416. MUL_MAT(type_a=q4_K,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 854 runs - 1176.20 us/run - 7.63 GFLOP/run -  6.49 TFLOPS
  417. MUL_MAT(type_a=q5_K,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 966 runs - 1045.79 us/run - 7.63 GFLOP/run -  7.30 TFLOPS
  418. MUL_MAT(type_a=q6_K,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 910 runs - 1099.93 us/run - 7.63 GFLOP/run -  6.94 TFLOPS
  419. MUL_MAT(type_a=iq2_xxs,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 826 runs - 1224.94 us/run - 7.63 GFLOP/run -  6.23 TFLOPS
  420. MUL_MAT(type_a=iq2_xs,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 798 runs - 1259.94 us/run - 7.63 GFLOP/run -  6.06 TFLOPS
  421. MUL_MAT(type_a=iq2_s,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 784 runs - 1278.98 us/run - 7.63 GFLOP/run -  5.97 TFLOPS
  422. MUL_MAT(type_a=iq3_xxs,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 812 runs - 1249.16 us/run - 7.63 GFLOP/run -  6.11 TFLOPS
  423. MUL_MAT(type_a=iq1_s,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 896 runs - 1121.19 us/run - 7.63 GFLOP/run -  6.81 TFLOPS
  424. MUL_MAT(type_a=iq4_nl,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 938 runs - 1071.12 us/run - 7.63 GFLOP/run -  7.13 TFLOPS
  425. MUL_MAT(type_a=iq3_s,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 826 runs - 1228.09 us/run - 7.63 GFLOP/run -  6.22 TFLOPS
  426. MUL_MAT(type_a=iq4_xs,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 952 runs - 1065.57 us/run - 7.63 GFLOP/run -  7.16 TFLOPS
  427. Backend ROCm0: OK
  428. Backend 2/4: ROCm1
  429. Skipping
  430. Backend 3/4: ROCm2
  431. Skipping
  432. Backend 4/4: CPU
  433. Skipping
  434. 4/4 backends passed
  435. OK
  436.  
  437. Testing 4 devices
  438.  
  439. Backend 1/4: ROCm0
  440. Device description: AMD Radeon Graphics
  441. Device memory: 32752 MB (32732 MB free)
  442.  
  443. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=1,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 9372 runs - 114.59 us/run - 117.44 MFLOP/run -  1.02 TFLOPS
  444. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=2,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 8520 runs - 119.64 us/run - 234.88 MFLOP/run -  1.96 TFLOPS
  445. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=3,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 7100 runs - 143.65 us/run - 352.32 MFLOP/run -  2.45 TFLOPS
  446. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=4,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5751 runs - 175.52 us/run - 469.76 MFLOP/run -  2.68 TFLOPS
  447. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=5,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 5472 runs - 187.66 us/run - 587.20 MFLOP/run -  3.13 TFLOPS
  448. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=8,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 3317 runs - 310.91 us/run - 939.52 MFLOP/run -  3.02 TFLOPS
  449. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1026 runs - 1003.14 us/run - 1.88 GFLOP/run -  1.87 TFLOPS
  450. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=24,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2052 runs - 490.74 us/run - 2.82 GFLOP/run -  5.74 TFLOPS
  451. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1782 runs - 565.97 us/run - 3.76 GFLOP/run -  6.64 TFLOPS
  452. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=40,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1540 runs - 656.43 us/run - 4.70 GFLOP/run -  7.16 TFLOPS
  453. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=48,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1350 runs - 746.86 us/run - 5.64 GFLOP/run -  7.55 TFLOPS
  454. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1064 runs - 941.23 us/run - 7.52 GFLOP/run -  7.99 TFLOPS
  455. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=512,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 186 runs - 5407.06 us/run - 60.13 GFLOP/run -  11.12 TFLOPS
  456. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=16,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 2322 runs - 439.88 us/run - 1.88 GFLOP/run -  4.27 TFLOPS
  457. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 663.58 us/run - 3.76 GFLOP/run -  5.66 TFLOPS
  458. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 924 runs - 1087.46 us/run - 7.52 GFLOP/run -  6.91 TFLOPS
  459. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=129,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 553 runs - 1822.57 us/run - 15.15 GFLOP/run -  8.31 TFLOPS
  460. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=145,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 438 runs - 2283.88 us/run - 17.03 GFLOP/run -  7.46 TFLOPS
  461. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=161,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 450 runs - 2239.64 us/run - 18.91 GFLOP/run -  8.44 TFLOPS
  462. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=177,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 350 runs - 2887.50 us/run - 20.79 GFLOP/run -  7.20 TFLOPS
  463. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=193,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 310 runs - 3263.45 us/run - 22.67 GFLOP/run -  6.95 TFLOPS
  464. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=209,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 285 runs - 3528.17 us/run - 24.55 GFLOP/run -  6.96 TFLOPS
  465. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=225,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 296 runs - 3389.91 us/run - 26.42 GFLOP/run -  7.79 TFLOPS
  466. MUL_MAT(type_a=q8_0,type_b=f32,m=4096,n=241,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 284 runs - 3549.82 us/run - 28.30 GFLOP/run -  7.97 TFLOPS
  467. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=129,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 553 runs - 1824.90 us/run - 15.39 GFLOP/run -  8.43 TFLOPS
  468. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=145,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 546 runs - 1833.77 us/run - 17.29 GFLOP/run -  9.43 TFLOPS
  469. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=161,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 438 runs - 2291.92 us/run - 19.20 GFLOP/run -  8.38 TFLOPS
  470. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=177,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 335 runs - 3016.82 us/run - 21.11 GFLOP/run -  7.00 TFLOPS
  471. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=193,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 345 runs - 2935.85 us/run - 23.02 GFLOP/run -  7.84 TFLOPS
  472. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=209,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 330 runs - 3067.43 us/run - 24.93 GFLOP/run -  8.13 TFLOPS
  473. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=225,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 312 runs - 3221.57 us/run - 26.84 GFLOP/run -  8.33 TFLOPS
  474. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=241,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 296 runs - 3378.45 us/run - 28.75 GFLOP/run -  8.51 TFLOPS
  475. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=129,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 560 runs - 1797.81 us/run - 15.15 GFLOP/run -  8.43 TFLOPS
  476. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=145,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 552 runs - 1830.90 us/run - 17.03 GFLOP/run -  9.30 TFLOPS
  477. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=161,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 438 runs - 2295.57 us/run - 18.91 GFLOP/run -  8.24 TFLOPS
  478. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=177,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 330 runs - 3064.02 us/run - 20.79 GFLOP/run -  6.79 TFLOPS
  479. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=193,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 335 runs - 2989.98 us/run - 22.67 GFLOP/run -  7.58 TFLOPS
  480. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=209,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 325 runs - 3121.33 us/run - 24.55 GFLOP/run -  7.87 TFLOPS
  481. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=225,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 308 runs - 3275.34 us/run - 26.43 GFLOP/run -  8.07 TFLOPS
  482. MUL_MAT(type_a=q8_0,type_b=f32,m=4097,n=241,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 292 runs - 3447.90 us/run - 28.31 GFLOP/run -  8.21 TFLOPS
  483. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=32,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 1512 runs - 667.02 us/run - 3.82 GFLOP/run -  5.72 TFLOPS
  484. MUL_MAT(type_a=q8_0,type_b=f32,m=4160,n=64,k=14336,bs=[1,1],nr=[1,1],per=[0,1,2,3],k_v=0,o=1,src_overlap=0): 938 runs - 1077.04 us/run - 7.63 GFLOP/run -  7.09 TFLOPS
  485. Backend ROCm0: OK
  486. Backend 2/4: ROCm1
  487. Skipping
  488. Backend 3/4: ROCm2
  489. Skipping
  490. Backend 4/4: CPU
  491. Skipping
  492. 4/4 backends passed
  493. OK
  494.  
Advertisement
Add Comment
Please, Sign In to add comment