llama 7B mostly
158 records across 21 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M5 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 104.03 | 990.53 | – | – | tg128 | 2026-03-31 | community | – | source |
| M5 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 102.93 | 987.1 | – | – | tg128 | 2026-03-09 | community | – | source |
| M2 Ultra | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 95.34 | 1240.61 | – | – | tg128 | 2023-12-04 | community | – | source |
| M2 Ultra | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 94.27 | 1238.48 | – | – | tg128 | 2023-11-22 | community | – | source |
| M3 Ultra 24+8 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 92.14 | 1471.24 | – | – | tg128 | 2025-03-23 | community | – | source |
| M3 Ultra | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 91.93 | 1473.76 | – | – | tg128 | 2025-04-22 | community | – | source |
| M2 Ultra | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 88.64 | 1013.81 | – | – | tg128 | 2023-11-23 | community | – | source |
| M3 Ultra 20+8 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 88.4 | 1073.09 | – | – | tg128 | 2025-03-14 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 83.73 | 1030.04 | – | – | tg128 | 2024-01-11 | community | – | source |
| M4 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 83.06 | 885.68 | – | – | tg128 | 2024-12-04 | community | – | source |
| M4 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 82.52 | 746.09 | – | – | tg128 | 2024-12-17 | community | – | source |
| M4 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 82.36 | 884.59 | – | – | tg128 | 2025-02-19 | community | – | source |
| M1 Ultra | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 74.93 | 772.24 | – | – | tg128 | 2023-12-01 | community | – | source |
| M3 | llama 7B mostly | Q4_0 | llama.cpp (CUDA) | 73.38 | 1816.45 | – | – | tg128 | 2024-01-21 | community | – | source |
| M3 | llama 7B mostly | Q4_0 | llama.cpp (CUDA) | 69.83 | 1956.45 | – | – | tg128 | 2024-01-21 | community | – | source |
| M3 | llama 7B mostly | Q4_0 | llama.cpp (CUDA) | 69.49 | 1087.59 | – | – | tg128 | 2024-01-11 | community | – | source |
| M4 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 69.24 | 715.74 | – | – | tg128 | 2025-05-20 | community | – | source |
| M2 Ultra | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 67.36 | 1250.86 | – | – | tg128 | 2023-12-04 | community | – | source |
| M2 Ultra | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 66.64 | 1248.59 | – | – | tg128 | 2023-11-22 | community | – | source |
| M3 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 66.31 | 759.7 | – | – | tg128 | 2023-11-24 | community | – | source |
| M2 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 65.95 | 671.31 | – | – | tg128 | 2023-11-22 | community | – | source |
| M3 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 65.85 | 690.99 | – | – | tg128 | 2023-11-22 | community | – | source |
| M5 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 65.11 | 1054.83 | – | – | tg128 | 2026-03-31 | community | – | source |
| M5 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 64.61 | 1051.59 | – | – | tg128 | 2026-03-09 | community | – | source |
| M2 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 64.55 | 669.51 | – | – | tg128 | 2023-11-25 | community | – | source |
| M3 Ultra | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 64.16 | 1488.84 | – | – | tg128 | 2025-04-22 | community | – | source |
| M3 Ultra 24+8 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 63.93 | 1487.51 | – | – | tg128 | 2025-03-23 | community | – | source |
| M3 Ultra 20+8 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 63.55 | 1085.76 | – | – | tg128 | 2025-03-14 | community | – | source |
| M2 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 63.29 | 664.07 | – | – | tg128 | 2023-11-25 | community | – | source |
| M2 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 62.98 | 668.28 | – | – | tg128 | 2023-11-26 | community | – | source |
| M2 Ultra | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 62.14 | 1003.16 | – | – | tg128 | 2023-11-23 | community | – | source |
| M1 Max | llama 7B mostly | Q4_0 | llama.cpp (CUDA) | 61.39 | 1945.87 | – | – | tg128 | 2023-12-17 | community | – | source |
| M1 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 61.19 | 530.06 | – | – | tg128 | 2023-11-25 | community | – | source |
| M2 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 60.99 | 537.6 | – | – | tg128 | 2023-11-29 | community | – | source |
| M1 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 60.96 | 529.4 | – | – | tg128 | 2023-12-17 | community | – | source |
| M1 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 60.4 | 527.76 | – | – | tg128 | 2023-12-21 | community | – | source |
| M5 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 60.04 | 403.19 | – | – | tg128 | 2026-04-14 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 59.87 | 1042.95 | – | – | tg128 | 2024-01-11 | community | – | source |
| M1 Max | llama 7B mostly | Q4_0 | llama.cpp (CUDA) | 57.87 | 1183.28 | – | – | tg1024 | 2023-12-17 | community | – | source |
| M1 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 57.39 | 466.05 | – | – | tg1024 | 2023-12-17 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 56.85 | 567.59 | – | – | tg128 | 2023-12-22 | community | – | source |
| M1 Ultra | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 55.69 | 783.45 | – | – | tg128 | 2023-12-01 | community | – | source |
| M1 Max | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 54.61 | 400.26 | – | – | tg128 | 2023-12-02 | community | contradiction | source |
| M4 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 54.05 | 891.94 | – | – | tg128 | 2024-12-04 | community | – | source |
| M4 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 53.75 | 891.32 | – | – | tg128 | 2025-02-19 | community | – | source |
| M4 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 53.06 | 852.47 | – | – | tg128 | 2024-12-17 | community | – | source |
| M3 | llama 7B mostly | Q8_0 | llama.cpp (CUDA) | 50.75 | 1937.77 | – | – | tg128 | 2024-01-21 | community | – | source |
| M4 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 50.74 | 439.78 | – | – | tg128 | 2024-11-08 | community | – | source |
| M3 | llama 7B mostly | Q8_0 | llama.cpp (CUDA) | 50.47 | 1049.96 | – | – | tg128 | 2024-01-11 | community | – | source |
| M4 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 49.64 | 364.06 | – | – | tg128 | 2024-12-07 | community | – | source |
| M4 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 43.8 | 720.38 | – | – | tg128 | 2025-05-20 | community | – | source |
| M3 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 43.0 | 749.37 | – | – | tg128 | 2023-11-22 | community | – | source |
| M3 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 42.75 | 757.64 | – | – | tg128 | 2023-11-24 | community | – | source |
| M3 | llama 7B mostly | Q8_0 | llama.cpp (CUDA) | 42.43 | 1688.37 | – | – | tg128 | 2024-01-21 | community | – | source |
| M3 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 42.37 | 707.07 | – | – | tg128 | 2023-12-28 | community | – | source |
| M3 Ultra 20+8 | llama 7B mostly | F16 | llama.cpp (Metal) | 42.24 | 1121.8 | – | – | tg128 | 2025-03-14 | community | – | source |
| M2 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 41.83 | 677.91 | – | – | tg128 | 2023-11-22 | community | – | source |
| M2 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 41.79 | 674.5 | – | – | tg128 | 2023-11-25 | community | – | source |
| M2 Ultra | llama 7B mostly | F16 | llama.cpp (Metal) | 41.34 | 1408.01 | – | – | tg128 | 2023-12-04 | community | – | source |
| M2 Ultra | llama 7B mostly | F16 | llama.cpp (Metal) | 41.02 | 1401.85 | – | – | tg128 | 2023-11-22 | community | – | source |
| M2 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 40.85 | 671.33 | – | – | tg128 | 2023-11-25 | community | – | source |
| M2 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 40.67 | 674.37 | – | – | tg128 | 2023-11-26 | community | – | source |
| M1 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 40.2 | 537.37 | – | – | tg128 | 2023-11-25 | community | – | source |
| M1 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 40.17 | 537.23 | – | – | tg128 | 2023-12-17 | community | – | source |
| M3 Ultra | llama 7B mostly | F16 | llama.cpp (Metal) | 40.1 | 1527.74 | – | – | tg128 | 2025-04-22 | community | – | source |
| M2 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 39.97 | 540.15 | – | – | tg128 | 2023-11-29 | community | – | source |
| M2 Ultra | llama 7B mostly | F16 | llama.cpp (Metal) | 39.86 | 1128.59 | – | – | tg128 | 2023-11-23 | community | – | source |
| M3 Ultra 24+8 | llama 7B mostly | F16 | llama.cpp (Metal) | 39.78 | 1538.34 | – | – | tg128 | 2025-03-23 | community | – | source |
| M2 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 39.71 | 342.61 | – | – | tg128 | 2023-12-28 | community | – | source |
| M1 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 39.6 | 536.5 | – | – | tg128 | 2023-12-21 | community | – | source |
| M2 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 38.86 | 341.19 | – | – | tg128 | 2023-12-11 | community | – | source |
| M1 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 38.5 | 471.64 | – | – | tg1024 | 2023-12-17 | community | – | source |
| M5 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 38.06 | 1032.18 | – | – | tg128 | 2026-03-31 | community | – | source |
| M2 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 37.87 | 294.24 | – | – | tg128 | 2023-11-24 | community | – | source |
| M1 Max | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 37.81 | 405.87 | – | – | tg128 | 2023-12-02 | community | – | source |
| M5 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 37.58 | 1018.3 | – | – | tg128 | 2026-03-09 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | F16 | llama.cpp (Metal) | 37.01 | 1168.89 | – | – | tg128 | 2024-01-11 | community | – | source |
| M1 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 36.41 | 266.25 | – | – | tg128 | 2023-11-22 | community | – | source |
| M1 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 36.34 | 233.49 | – | – | tg128 | 2024-01-02 | community | – | source |
| M1 Ultra | llama 7B mostly | F16 | llama.cpp (Metal) | 36.31 | 1158.95 | – | – | tg128 | 2024-01-10 | community | – | source |
| M5 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 35.86 | 431.14 | – | – | tg128 | 2026-04-14 | community | – | source |
| M1 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 35.45 | 233.01 | – | – | tg128 | 2023-12-17 | community | – | source |
| M3 | llama 7B mostly | F16 | llama.cpp (CUDA) | 35.42 | 925.62 | – | – | tg128 | 2024-01-21 | community | – | source |
| M3 | llama 7B mostly | F16 | llama.cpp (CUDA) | 35.28 | 962.94 | – | – | tg128 | 2024-01-11 | community | – | source |
| M1 Max | llama 7B mostly | Q8_0 | llama.cpp (CUDA) | 35.02 | 1873.64 | – | – | tg128 | 2023-12-17 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 34.3 | 566.4 | – | – | tg128 | 2023-12-22 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 34.3 | 566.4 | – | – | tg128 | 2024-01-11 | community | contradiction | source |
| M1 Ultra | llama 7B mostly | F16 | llama.cpp (Metal) | 33.92 | 875.81 | – | – | tg128 | 2023-12-01 | community | – | source |
| M1 Max | llama 7B mostly | Q8_0 | llama.cpp (CUDA) | 33.86 | 1157.46 | – | – | tg1024 | 2023-12-17 | community | – | source |
| M4 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 31.64 | 922.83 | – | – | tg128 | 2024-12-04 | community | – | source |
| M4 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 31.61 | 923.55 | – | – | tg128 | 2024-12-17 | community | – | source |
| M4 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 31.56 | 920.48 | – | – | tg128 | 2025-02-19 | community | – | source |
| M3 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 30.74 | 341.67 | – | – | tg128 | 2023-12-09 | community | – | source |
| M4 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 30.69 | 449.62 | – | – | tg128 | 2024-11-08 | community | – | source |
| M3 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 30.65 | 269.49 | – | – | tg128 | 2023-11-25 | community | – | source |
| M1 Pro | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 30.6 | 205.41 | – | – | tg128 | 2023-12-17 | community | – | source |
| M4 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 30.54 | 367.13 | – | – | tg128 | 2024-12-07 | community | – | source |
| M3 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 30.36 | 338.37 | – | – | tg128 | 2023-12-01 | community | – | source |
| M5 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 29.62 | 247.68 | – | – | tg128 | 2025-10-30 | community | – | source |
| M4 | llama 7B mostly | F16 | llama.cpp (Metal) | 25.58 | 747.59 | – | – | tg128 | 2025-05-20 | community | – | source |
| M3 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 25.27 | 794.26 | – | – | tg128 | 2023-11-22 | community | – | source |
| M3 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 25.14 | 784.83 | – | – | tg128 | 2023-12-28 | community | – | source |
| M3 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 25.09 | 779.17 | – | – | tg128 | 2023-11-24 | community | – | source |
| M2 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 24.65 | 755.67 | – | – | tg128 | 2023-11-22 | community | – | source |
| M2 | llama 7B mostly | F16 | llama.cpp (Metal) | 24.54 | 747.99 | – | – | tg128 | 2023-11-26 | community | – | source |
| M2 | llama 7B mostly | F16 | llama.cpp (Metal) | 24.31 | 754.39 | – | – | tg128 | 2023-11-25 | community | – | source |
| M2 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 24.16 | 600.46 | – | – | tg128 | 2023-11-29 | community | – | source |
| M4 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 24.11 | 221.29 | – | – | tg128 | 2024-11-10 | community | – | source |
| M3 | llama 7B mostly | F16 | llama.cpp (CUDA) | 23.93 | 3022.37 | – | – | tg128 | 2024-01-21 | community | – | source |
| M2 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 23.24 | 335.9 | – | – | tg128 | 2023-12-28 | community | – | source |
| M1 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 23.03 | 599.53 | – | – | tg128 | 2023-11-25 | community | – | source |
| M2 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 23.01 | 344.5 | – | – | tg128 | 2023-12-11 | community | – | source |
| M1 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 22.96 | 596.85 | – | – | tg128 | 2023-12-17 | community | – | source |
| M1 | llama 7B mostly | F16 | llama.cpp (Metal) | 22.83 | 598.96 | – | – | tg128 | 2023-12-21 | community | – | source |
| M2 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 22.7 | 288.46 | – | – | tg128 | 2023-11-24 | community | – | source |
| M1 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 22.55 | 453.03 | – | – | tg128 | 2023-12-02 | community | – | source |
| M1 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 22.51 | 237.0 | – | – | tg128 | 2024-01-02 | community | – | source |
| M1 Max | llama 7B mostly | F16 | llama.cpp (Metal) | 22.41 | 516.68 | – | – | tg1024 | 2023-12-17 | community | – | source |
| M1 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 22.34 | 270.37 | – | – | tg128 | 2023-11-22 | community | – | source |
| M1 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 22.16 | 236.72 | – | – | tg128 | 2023-12-17 | community | – | source |
| M2 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 21.91 | 179.57 | – | – | tg128 | 2023-11-22 | community | – | source |
| M2 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 21.7 | 145.91 | – | – | tg128 | 2024-01-07 | community | – | source |
| M3 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 21.34 | 186.75 | – | – | tg128 | 2023-11-27 | community | – | source |
| M1 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 20.55 | 201.58 | – | – | tg128 | 2023-12-17 | community | – | source |
| M5 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 20.22 | 423.3 | – | – | tg128 | 2026-04-14 | community | – | source |
| M3 Max 10+4 | llama 7B mostly | F16 | llama.cpp (Metal) | 19.54 | 589.41 | – | – | tg128 | 2023-12-22 | community | – | source |
| M1 Max | llama 7B mostly | F16 | llama.cpp (CUDA) | 19.45 | 2757.29 | – | – | tg128 | 2023-12-17 | community | – | source |
| M3 | llama 7B mostly | Q4_0 | llama.cpp (CUDA) | 19.23 | 671.1 | – | – | tg128 | 2024-01-11 | community | contradiction outlier | source |
| M3 | llama 7B mostly | Q8_0 | llama.cpp (CUDA) | 19.11 | 637.51 | – | – | tg128 | 2024-01-11 | community | contradiction | source |
| M1 Max | llama 7B mostly | F16 | llama.cpp (CUDA) | 19.1 | 1453.27 | – | – | tg1024 | 2023-12-17 | community | – | source |
| M3 | llama 7B mostly | F16 | llama.cpp (CUDA) | 18.46 | 648.07 | – | – | tg128 | 2024-01-11 | community | contradiction | source |
| M3 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 17.53 | 344.66 | – | – | tg128 | 2023-12-09 | community | – | source |
| M3 Pro | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 17.44 | 272.11 | – | – | tg128 | 2023-11-25 | community | – | source |
| M4 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 17.19 | 381.14 | – | – | tg128 | 2024-12-07 | community | – | source |
| M4 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 17.18 | 464.48 | – | – | tg128 | 2024-11-08 | community | – | source |
| M5 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 16.62 | 264.15 | – | – | tg128 | 2025-10-30 | community | – | source |
| M3 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 16.46 | 339.31 | – | – | tg128 | 2023-12-01 | community | – | source |
| M1 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 14.19 | 107.81 | – | – | tg128 | 2023-12-15 | community | – | source |
| M1 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 14.15 | 117.96 | – | – | tg128 | 2023-12-02 | community | – | source |
| M1 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 14.13 | 115.67 | – | – | tg128 | 2025-05-20 | community | – | source |
| M4 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 13.54 | 223.64 | – | – | tg128 | 2024-11-10 | community | – | source |
| M2 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 13.06 | 384.38 | – | – | tg128 | 2023-12-11 | community | – | source |
| M2 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 12.97 | 373.97 | – | – | tg128 | 2023-12-28 | community | – | source |
| M1 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 12.75 | 302.14 | – | – | tg128 | 2023-11-22 | community | – | source |
| M1 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 12.75 | 262.65 | – | – | tg128 | 2024-01-02 | community | – | source |
| M2 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 12.47 | 312.65 | – | – | tg128 | 2023-11-24 | community | – | source |
| M3 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 12.27 | 187.52 | – | – | tg128 | 2023-11-27 | community | – | source |
| M2 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 12.21 | 181.4 | – | – | tg128 | 2023-11-22 | community | – | source |
| M2 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 12.18 | 147.27 | – | – | tg128 | 2024-01-07 | community | – | source |
| M3 Pro | llama 7B mostly | F16 | llama.cpp (Metal) | 9.89 | 357.45 | – | – | tg128 | 2023-12-09 | community | – | source |
| M5 | llama 7B mostly | F16 | llama.cpp (Metal) | 9.43 | 257.79 | – | – | tg128 | 2025-10-30 | community | – | source |
| M2 | llama 7B mostly | Q4_0 | llama.cpp (Metal) | 8.61 | 85.24 | – | – | tg128 | 2023-12-05 | community | contradiction | source |
| M1 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 7.92 | 108.21 | – | – | tg128 | 2023-12-15 | community | – | source |
| M1 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 7.91 | 117.25 | – | – | tg128 | 2023-12-02 | community | – | source |
| M1 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 7.69 | 121.73 | – | – | tg128 | 2025-05-20 | community | – | source |
| M4 | llama 7B mostly | F16 | llama.cpp (Metal) | 7.43 | 230.18 | – | – | tg128 | 2024-11-10 | community | – | source |
| M2 | llama 7B mostly | F16 | llama.cpp (Metal) | 6.72 | 201.34 | – | – | tg128 | 2023-11-22 | community | – | source |
| M2 | llama 7B mostly | Q8_0 | llama.cpp (Metal) | 5.31 | 76.9 | – | – | tg128 | 2023-12-05 | community | contradiction | source |