Qwen3.5-9B
805 records across 35 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| H200 | Qwen3.5-9B | BF16 | vLLM | 2069.343 | – | – | – | – | 2026-10-01 | community | – | source |
| H200 | Qwen3.5-9B | BF16 | vLLM | 1859.655 | – | – | – | – | 2026-10-01 | community | – | source |
| H200 | Qwen3.5-9B | BF16 | vLLM | 1536.536 | – | – | – | – | 2026-10-01 | community | – | source |
| H200 | Qwen3.5-9B | BF16 | vLLM | 1426.638 | – | – | – | – | 2026-10-01 | community | – | source |
| Radeon PRO W7900 | Qwen3.5-9B 9B | Q4_K_M | llama.cpp (Vulkan) | 218 | – | – | – | – | 2026-08 | community | – | source |
| 5090 | Qwen3.5-9B | Q4_K_S MTP | LM Studio | 185 | – | – | – | – | 2026-07-19 | community | – | source |
| RTX 3090 | Qwen3.5-9B | IQ1_S | llama.cpp (CUDA) | 168.1 | – | – | – | 16384 | 2026-10-04 | community | – | source |
| RTX 3090 | Qwen3.5-9B | IQ2_XXS | llama.cpp (CUDA) | 156.4 | – | – | – | 16384 | 2026-10-04 | community | – | source |
| RTX 3090 | Qwen3.5-9B | IQ2_M | llama.cpp (CUDA) | 145.2 | – | – | – | 16384 | 2026-10-04 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 142.5 | 2766.0 | – | – | 1024 | 2026-10-14 | community | – | source |
| RTX 3090 | Qwen3.5-9B | IQ3_XXS | llama.cpp (CUDA) | 138.7 | – | – | – | 16384 | 2026-10-04 | community | – | source |
| 5090 | Qwen3.5-9B | Q4_K_S | LM Studio | 130 | – | – | – | – | 2026-07-19 | community | – | source |
| RX 7800 XT | Qwen3.5-9B 9B | Q4_K_M | llama.cpp (Vulkan) | 124 | – | – | – | – | 2026-08 | community | – | source |
| RTX 3090 | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | 112.5 | – | – | – | 16384 | 2026-10-04 | community | – | source |
| M3 Ultra | Qwen3.5-9B | 4bit | oMLX | 112.1 | 1530.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M3 Ultra | Qwen3.5-9B | 4bit | oMLX | 111.3 | 1228.0 | – | – | 4096 | 2026-10-07 | community | – | source |
| M3 Ultra | Qwen3.5-9B | 4bit | oMLX | 109.6 | 1653.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M3 Ultra (256 GB) | Qwen3.5-9B | 4bit | MLX | 106 | – | – | – | – | – | sourced | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 105.0 | 3034.0 | – | – | 1024 | 2026-10-08 | community | contradiction | source |
| 2x EVGA RTX 3080 Ti 12GB | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | 104.8 | 1423 | – | – | 262144 | 2026-04-02 | community | – | source |
| 2x EVGA RTX 3080 Ti 12GB | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | 104.7 | 2027 | – | – | 131072 | 2026-04-02 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 104.5 | 3146.0 | – | – | 1024 | 2026-10-10 | community | contradiction | source |
| 2x EVGA RTX 3080 Ti 12GB | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | 104.4 | 1827 | – | – | 163840 | 2026-04-02 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 104.2 | 3129.0 | – | – | 1024 | 2026-10-10 | community | contradiction | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 103.5 | 2905.0 | – | – | 1024 | 2026-10-15 | community | contradiction | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 103.2 | 3567.0 | – | – | 4096 | 2026-10-08 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 102.6 | 3907.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 101.2 | 2896.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 101.0 | 3472.0 | – | – | 8192 | 2026-10-10 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 100.9 | 3553.0 | – | – | 8192 | 2026-10-08 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 100.8 | 3037.0 | – | – | 1024 | 2026-10-08 | community | contradiction | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 99.1 | 3500.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 99.1 | 766.3 | – | – | 1024 | 2026-10-16 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 98.9 | 4545.0 | – | – | 8192 | 2026-10-10 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 98.9 | 771.2 | – | – | 1024 | 2026-10-16 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 97.0 | 722.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 97.0 | 722.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.7 | 727.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.7 | 727.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.7 | 727.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.7 | 727.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.7 | 727.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.7 | 727.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| Radeon PRO W7900 | Qwen3.5-9B 9B | Q4_K_M | llama.cpp (Vulkan) | 96 | – | – | – | – | 2026-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 96.0 | 723.5 | – | – | 4096 | 2026-10-16 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 95.7 | 3433.0 | – | – | 16384 | 2026-10-08 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 95.7 | 4563.0 | – | – | 16384 | 2026-10-10 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 95.7 | 3314.0 | – | – | 16384 | 2026-10-10 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 95.5 | 725.2 | – | – | 4096 | 2026-10-16 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.7 | 786.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.7 | 786.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 93.7 | 3401.0 | – | – | 16384 | 2026-10-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 789.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 789.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 789.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 789.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 789.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 789.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 93.2 | 713.2 | – | – | 8192 | 2026-10-16 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 92.9 | 712.4 | – | – | 8192 | 2026-10-16 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 91.9 | 361.1 | – | – | 32768 | 2026-10-13 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 90.7 | 889.7 | – | – | 1024 | 2026-10-09 | community | – | source |
| RTX PRO 4000 Blackwell | Qwen3.5-9B 9B | Q4_K_M | llama.cpp (CUDA) | 90 | – | – | – | – | 2026-08 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 88.8 | 982.0 | – | – | 4096 | 2026-10-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 88.0 | 3189.0 | – | – | 32768 | 2026-10-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.6 | 792.6 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.6 | 792.6 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.6 | 792.6 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.6 | 792.6 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.6 | 792.6 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.6 | 792.6 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 87.4 | 758.3 | – | – | 4096 | 2026-10-14 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 86.9 | 3167.0 | – | – | 32768 | 2026-10-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 86.9 | 2951.0 | – | – | 32768 | 2026-10-10 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 86.8 | 861.6 | – | – | 1024 | 2026-10-14 | community | contradiction | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 86.7 | 4193.0 | – | – | 32768 | 2026-10-10 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 86.6 | 962.7 | – | – | 8192 | 2026-10-09 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 84.2 | 950.7 | – | – | 8192 | 2026-10-09 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 82.1 | 890.7 | – | – | 4096 | 2026-10-10 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 81.5 | 923.7 | – | – | 16384 | 2026-10-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 79.0 | 253.9 | – | – | 32768 | 2026-10-13 | community | contradiction | source |
| RTX 3090 | Qwen3.5-9B | unspecified | llama.cpp (CUDA) | 78.2 | – | – | – | 16384 | 2026-10-04 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 76.6 | 774.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 76.6 | 774.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 76.6 | 774.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 76.6 | 774.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 76.6 | 774.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 76.6 | 774.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 74.5 | 2807.0 | – | – | 65536 | 2026-10-06 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 74.5 | 2778.0 | – | – | 65536 | 2026-10-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 74.3 | 775.8 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 74.3 | 775.8 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 74.3 | 775.8 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 74.3 | 775.8 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 74.3 | 775.8 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 74.3 | 775.8 | – | – | 1024 | 2026-03-08 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 73.7 | 2760.0 | – | – | 65536 | 2026-10-09 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 73.6 | 851.8 | – | – | 32768 | 2026-10-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 73.4 | 3369.0 | – | – | 65536 | 2026-10-10 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 72.3 | 869.5 | – | – | 16384 | 2026-10-10 | community | contradiction | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 72.3 | 2399.0 | – | – | 65536 | 2026-10-10 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 72.1 | 783.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 72.1 | 783.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 72.1 | 783.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 72.1 | 783.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 72.1 | 783.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 72.1 | 783.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Max | Qwen3.5-9B | 4bit | oMLX | 72.1 | 794.8 | – | – | 1024 | 2026-10-08 | community | – | source |
| RX 7800 XT | Qwen3.5-9B 9B | Q4_K_M | llama.cpp (Vulkan) | 70 | – | – | – | – | 2026-08 | community | – | source |
| M3 Max | Qwen3.5-9B | 4bit | oMLX | 70.0 | 849.5 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 69.8 | 742.9 | – | – | 1024 | 2026-10-12 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 68.7 | 776.7 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 68.7 | 776.7 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 68.7 | 776.7 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 68.7 | 776.7 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 68.7 | 776.7 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 68.7 | 776.7 | – | – | 8192 | 2026-03-08 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 67.9 | 751.4 | – | – | 4096 | 2026-10-12 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 66.7 | 4400.0 | – | – | 4096 | 2026-10-08 | community | contradiction | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 66.0 | 627.9 | – | – | 8192 | 2026-10-12 | community | – | source |
| M5 Max | Qwen3.5-9B | 6bit | oMLX | 64.4 | 2872.0 | – | – | 4096 | 2026-10-12 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 64.3 | 794.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 64.3 | 794.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 64.3 | 794.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 64.3 | 794.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 64.3 | 794.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 64.3 | 794.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 63.1 | 441.2 | – | – | 16384 | 2026-10-13 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 62.8 | 758.8 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 62.8 | 758.8 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 62.8 | 758.8 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 62.8 | 758.8 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 62.8 | 758.8 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 62.8 | 758.8 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 62.6 | 804.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 62.6 | 804.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 62.6 | 804.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 62.6 | 804.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 62.6 | 804.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 62.6 | 804.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 62.3 | 404.6 | – | – | 1024 | 2026-10-09 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 62.1 | 735.7 | – | – | 65536 | 2026-10-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 61.9 | 738.1 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 61.9 | 738.1 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 61.9 | 738.1 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 61.9 | 738.1 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 61.9 | 738.1 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 61.9 | 738.1 | – | – | 32768 | 2026-03-09 | community | – | source |
| RTX PRO 4000 Blackwell | Qwen3.5-9B 9B | Q4_K_M | llama.cpp (CUDA) | 61 | – | – | – | – | 2026-08 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 59.9 | 392.7 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 59.9 | 392.7 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 59.9 | 392.7 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 59.7 | 797.4 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 59.7 | 797.4 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 59.7 | 797.4 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 59.7 | 797.4 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 59.7 | 797.4 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 59.7 | 797.4 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 59.7 | 398.0 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 59.7 | 398.0 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 59.7 | 398.0 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 58.7 | 388.2 | – | – | 4096 | 2026-10-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 58.6 | 392.3 | – | – | 1024 | 2026-10-07 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 58.5 | 386.1 | – | – | 1024 | 2026-10-12 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 58.2 | 401.8 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 58.2 | 401.8 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 57.7 | 1758.0 | – | – | 1024 | 2026-10-07 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 57.0 | 2186.0 | – | – | 131072 | 2026-10-08 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 57.0 | 1409.0 | – | – | 1024 | 2026-10-15 | community | – | source |
| M1 Max | Qwen3.5-9B | oQ4e | oMLX | 56.7 | 382.2 | – | – | 1024 | 2026-10-08 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 56.7 | 2516.0 | – | – | 131072 | 2026-10-10 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 56.6 | 312.5 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 56.6 | 312.5 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 56.5 | 1788.0 | – | – | 4096 | 2026-10-07 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 56.5 | 380.4 | – | – | 4096 | 2026-10-12 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 56.2 | 1523.0 | – | – | 1024 | 2026-10-16 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 55.9 | 1485.0 | – | – | 4096 | 2026-10-17 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 55.4 | 1491.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M1 Max | Qwen3.5-9B | oQ4e | oMLX | 55.2 | 397.3 | – | – | 4096 | 2026-10-08 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 55.2 | 281.4 | – | – | 16384 | 2026-10-13 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 55.1 | 308.2 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 55.1 | 308.2 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 55.0 | 1822.0 | – | – | 8192 | 2026-10-07 | community | – | source |
| M5 Max | Qwen3.5-9B | 6bit | oMLX | 54.7 | 1843.0 | – | – | 16384 | 2026-10-12 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 54.7 | 1740.0 | – | – | 4096 | 2026-10-16 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 54.3 | 1475.0 | – | – | 8192 | 2026-10-17 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 54.0 | 778.7 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 54.0 | 778.7 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 54.0 | 778.7 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 54.0 | 778.7 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 54.0 | 778.7 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 54.0 | 778.7 | – | – | 16384 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 53.8 | 317.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 53.8 | 317.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 53.6 | 408.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 53.6 | 408.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 53.5 | 1954.0 | – | – | 131072 | 2026-10-10 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 53.2 | 311.9 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 53.2 | 311.9 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 52.6 | 724.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 52.6 | 724.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 52.6 | 724.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 52.6 | 724.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 52.6 | 724.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 52.6 | 724.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M4 Max | Qwen3.5-9B | 8bit | oMLX | 52.4 | 903.6 | – | – | 1024 | 2026-10-15 | community | – | source |
| M4 Max | Qwen3.5-9B | 8bit | oMLX | 52.1 | 878.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 Max | Qwen3.5-9B | 8bit | oMLX | 52.1 | 878.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 51.6 | 1759.0 | – | – | 16384 | 2026-10-07 | community | – | source |
| M4 Max | Qwen3.5-9B | 8bit | oMLX | 51.4 | 936.1 | – | – | 4096 | 2026-10-15 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 51.1 | 365.6 | – | – | 4096 | 2026-10-07 | community | – | source |
| M4 Max | Qwen3.5-9B | 8bit | oMLX | 50.9 | 890.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 Max | Qwen3.5-9B | 8bit | oMLX | 50.9 | 890.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 50.9 | 379.9 | – | – | 1024 | 2026-10-12 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 50.7 | 331.0 | – | – | 16384 | 2026-10-10 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 50.6 | 466.5 | – | – | 1024 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 50.5 | 373.5 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 50.5 | 380.2 | – | – | 1024 | 2026-10-15 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 50.2 | 462.2 | – | – | 1024 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | mxfp4 | oMLX | 50.1 | 374.8 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.8 | 379.1 | – | – | 4096 | 2026-10-12 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.5 | 478.3 | – | – | 4096 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.4 | 372.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.2 | 390.1 | – | – | 4096 | 2026-10-15 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.1 | 385.7 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.0 | 481.5 | – | – | 4096 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 49.0 | 383.0 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 48.9 | 445.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 48.9 | 445.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 48.9 | 362.2 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 48.8 | 378.7 | – | – | 1024 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 48.1 | 373.9 | – | – | 8192 | 2026-10-12 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 48.0 | 353.1 | – | – | 8192 | 2026-10-07 | community | – | source |
| M5 Max | Qwen3.5-9B | 8bit | oMLX | 47.8 | 2371.0 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 47.7 | 532.1 | – | – | 4096 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 47.7 | 367.9 | – | – | 1024 | 2026-10-16 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 47.6 | 447.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 47.6 | 447.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 47.6 | 280.7 | – | – | 1024 | 2026-10-15 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 47.3 | 284.7 | – | – | 1024 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 47.3 | 284.7 | – | – | 1024 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 47.3 | 284.7 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 47.3 | 379.9 | – | – | 4096 | 2026-10-16 | community | – | source |
| M4 Max | Qwen3.5-9B | 4bit | oMLX | 47.1 | 577.9 | – | – | 131072 | 2026-10-09 | community | – | source |
| M5 Max | Qwen3.5-9B | 8bit | oMLX | 47.0 | 2672.0 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 47.0 | 363.9 | – | – | 1024 | 2026-10-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 46.8 | 375.9 | – | – | 4096 | 2026-10-16 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 46.4 | 742.0 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 46.4 | 742.0 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 46.4 | 742.0 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 46.4 | 742.0 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 46.4 | 742.0 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 46.4 | 742.0 | – | – | 32768 | 2026-03-09 | community | – | source |
| M3 Pro | Qwen3.5-9B | oQ4e | oMLX | 46.2 | 339.1 | – | – | 16384 | 2026-10-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 46.2 | 361.4 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 46.2 | 375.9 | – | – | 4096 | 2026-10-16 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 46.0 | 375.4 | – | – | 4096 | 2026-10-08 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 45.9 | 276.2 | – | – | 4096 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 45.9 | 276.2 | – | – | 4096 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 45.9 | 276.2 | – | – | 4096 | 2026-10-06 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 45.8 | 1776.0 | – | – | 199680 | 2026-10-08 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 45.7 | 376.1 | – | – | 4096 | 2026-10-16 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 45.6 | 366.9 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 45.6 | 366.9 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 45.6 | 366.9 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 45.5 | 1545.0 | – | – | 199680 | 2026-10-10 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 45.3 | 331.7 | – | – | 16384 | 2026-10-07 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 45.1 | 355.8 | – | – | 16384 | 2026-10-12 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 45.0 | 373.0 | – | – | 8192 | 2026-10-16 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 44.9 | 1930.0 | – | – | 199680 | 2026-10-10 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 44.9 | 476.9 | – | – | 8192 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 44.8 | 344.4 | – | – | 16384 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 44.8 | 371.7 | – | – | 4096 | 2026-10-16 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 44.6 | 364.6 | – | – | 4096 | 2026-10-09 | community | – | source |
| AMD RX 5700 XT 8GB (gfx1010, RDNA1) | Qwen3.5-9B | Q4_K_M | llama.cpp (Vulkan) | 44.5 | 269.7 | – | – | 10890 | 2026-10 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 44.5 | 414.7 | – | – | 1024 | 2026-03-09 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 44.5 | 414.7 | – | – | 1024 | 2026-03-09 | community | contradiction | source |
| AMD RX 5700 XT 8GB (gfx1010, RDNA1) | Qwen3.5-9B | Q4_K_M | llama.cpp (Vulkan) | 44.3 | 151.6 | – | – | 10890 | 2026-10 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 44.3 | 665.7 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 44.3 | 665.7 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 44.3 | 665.7 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 44.3 | 665.7 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 44.3 | 665.7 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 44.3 | 665.7 | – | – | 65536 | 2026-03-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 44.3 | 367.8 | – | – | 16384 | 2026-10-11 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 44.2 | 363.8 | – | – | 4096 | 2026-10-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 44.2 | 361.3 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 43.6 | 358.2 | – | – | 16384 | 2026-10-13 | community | – | source |
| M5 Max | Qwen3.5-9B | 4bit | oMLX | 43.4 | 2299.0 | – | – | 1024 | 2026-10-08 | community | contradiction | source |
| M3 Pro | Qwen3.5-9B | oQ4e | oMLX | 43.3 | 351.9 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 43.2 | 411.9 | – | – | 4096 | 2026-03-09 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 43.2 | 411.9 | – | – | 4096 | 2026-03-09 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 43.2 | 368.7 | – | – | 8192 | 2026-10-16 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 42.1 | 295.0 | – | – | 32768 | 2026-10-07 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 42.1 | 438.8 | – | – | 16384 | 2026-10-16 | community | – | source |
| M1 Max | Qwen3.5-9B | OptiQ-4bit | oMLX | 41.7 | 289.3 | – | – | 4096 | 2026-10-09 | community | – | source |
| M1 Max | Qwen3.5-9B | OptiQ-4bit | oMLX | 41.6 | 287.7 | – | – | 4096 | 2026-10-09 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 41.6 | 346.0 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 41.6 | 346.0 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 41.6 | 346.0 | – | – | 8192 | 2026-10-11 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 41.5 | 343.0 | – | – | 16384 | 2026-10-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 41.1 | 589.4 | – | – | 1024 | 2026-10-14 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 40.4 | 359.8 | – | – | 16384 | 2026-10-08 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 40.4 | 1485.0 | – | – | 4096 | 2026-10-13 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 40.4 | 495.4 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 39.9 | 326.1 | – | – | 32768 | 2026-10-12 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 39.8 | 653.8 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 39.8 | 653.8 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 39.8 | 653.8 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 39.8 | 653.8 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 39.8 | 653.8 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 6bit | oMLX | 39.8 | 653.8 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 39.6 | 341.0 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 39.6 | 341.0 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 39.6 | 341.0 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 39.5 | 1430.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 39.1 | 328.0 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 39.1 | 328.0 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 Max | Qwen3.5-9B | 4bit | oMLX | 39.1 | 328.0 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 38.7 | 329.5 | – | – | 32768 | 2026-10-13 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 38.3 | 1465.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 38.3 | 376.4 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 38.2 | 372.6 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 37.8 | 374.8 | – | – | 1024 | 2026-10-13 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 37.7 | 250.2 | – | – | 32768 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 37.7 | 227.2 | – | – | 8192 | 2026-10-18 | community | – | source |
| RTX 4060 Ti | Qwen3.5-9B | Q6_K | Vulkan | 37.64 | 2430.82 | – | – | 512 | 2026-10-02 | community | – | source |
| RTX 4060 Ti | Qwen3.5-9B | Q6_K | Vulkan | 37.55 | 2407.05 | – | – | 512 | 2026-10-02 | community | – | source |
| RTX 4060 Ti | Qwen3.5-9B | Q6_K | Vulkan | 37.54 | 2314.17 | – | – | 512 | 2026-10-02 | community | – | source |
| RTX 4060 Ti | Qwen3.5-9B | Q6_K | Vulkan | 37.46 | 2320.02 | – | – | 512 | 2026-10-02 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 37.3 | 719.3 | – | – | 1024 | 2026-10-10 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 37.2 | 322.6 | – | – | 32768 | 2026-10-09 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 37.2 | 1407.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 37.2 | 332.8 | – | – | 32768 | 2026-10-16 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 36.9 | 363.0 | – | – | 1024 | 2026-10-16 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 36.6 | 216.2 | – | – | 1024 | 2026-10-06 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 36.4 | 223.4 | – | – | 1024 | 2026-10-08 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 36.3 | 830.3 | – | – | 4096 | 2026-10-10 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 36.2 | 295.2 | – | – | 8192 | 2026-10-13 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.9 | 316.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.9 | 316.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 35.8 | 276.7 | – | – | 1024 | 2026-10-13 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ4e | oMLX | 35.7 | 298.2 | – | – | 4096 | 2026-10-13 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.6 | 304.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.6 | 304.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 35.4 | 249.3 | – | – | 1024 | 2026-10-18 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 35.4 | 249.3 | – | – | 1024 | 2026-10-18 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.3 | 239.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.3 | 239.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 35.3 | 208.4 | – | – | 4096 | 2026-10-06 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 35.3 | 1542.0 | – | – | 4096 | 2026-10-13 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.1 | 239.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.1 | 239.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.1 | 239.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.1 | 239.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.1 | 239.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 35.1 | 239.3 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 35.1 | 858.6 | – | – | 8192 | 2026-10-10 | community | contradiction | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 35.0 | 223.0 | – | – | 1024 | 2026-10-18 | community | – | source |
| M3 Ultra | Qwen3.5-9B | unknown | oMLX | 34.9 | 928.9 | – | – | 65536 | 2026-10-08 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 34.7 | 319.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 34.7 | 321.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 34.7 | 319.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 34.7 | 321.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 34.7 | 230.6 | – | – | 4096 | 2026-10-08 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 34.6 | 1545.0 | – | – | 4096 | 2026-10-13 | community | contradiction | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 34.5 | 189.4 | – | – | 1024 | 2026-10-15 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 34.3 | 240.5 | – | – | 4096 | 2026-10-07 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 34.1 | 253.1 | – | – | 4096 | 2026-10-18 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 34.1 | 253.1 | – | – | 4096 | 2026-10-18 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 34.0 | 194.1 | – | – | 1024 | 2026-10-16 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 33.9 | 668.6 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 33.9 | 668.6 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 33.9 | 668.6 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 33.9 | 668.6 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 33.9 | 668.6 | – | – | 65536 | 2026-03-09 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 8bit | oMLX | 33.9 | 668.6 | – | – | 65536 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 33.9 | 188.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 33.5 | 193.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 33.5 | 193.0 | – | – | 1024 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.4 | 241.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.4 | 241.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 33.4 | 230.8 | – | – | 4096 | 2026-10-08 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.3 | 242.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.3 | 242.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.3 | 242.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.3 | 242.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.3 | 242.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 33.3 | 242.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 33.3 | 200.4 | – | – | 32768 | 2026-10-18 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 33.2 | 816.1 | – | – | 16384 | 2026-10-10 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 33.1 | 223.6 | – | – | 1024 | 2026-10-08 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | 33 | – | – | – | tg32 | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | 33 | – | – | – | tg32 | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | 33 | – | – | – | tg32 | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | 33 | – | – | – | tg32 | 2026-10 | community | – | source |
| M5 Pro | Qwen3.5-9B | 8bit | oMLX | 32.6 | 1218.0 | – | – | 1024 | 2026-10-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 32.5 | 188.7 | – | – | 4096 | 2026-10-06 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 32.5 | 188.7 | – | – | 4096 | 2026-10-06 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 32.3 | 158.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 32.3 | 158.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 32.3 | 188.5 | – | – | 4096 | 2026-10-16 | community | – | source |
| M5 Pro | Qwen3.5-9B | 8bit | oMLX | 32.2 | 1384.0 | – | – | 4096 | 2026-10-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 32.2 | 417.3 | – | – | 16384 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | oQ4 | oMLX | 32.0 | 221.1 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 31.9 | 280.1 | – | – | 65536 | 2026-10-12 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 31.7 | 180.0 | – | – | 8192 | 2026-10-16 | community | – | source |
| M5 Pro | Qwen3.5-9B | 8bit | oMLX | 31.7 | 1320.0 | – | – | 4096 | 2026-10-17 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 31.7 | 1129.0 | – | – | 4096 | 2026-10-18 | community | contradiction | source |
| M5 Pro | Qwen3.5-9B | 8bit | oMLX | 31.6 | 1412.0 | – | – | 8192 | 2026-10-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 31.5 | 159.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 31.5 | 159.3 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | OptiQ-4bit | oMLX | 31.1 | 1446.0 | – | – | 4096 | 2026-10-16 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 30.8 | 974.6 | – | – | 1024 | 2026-10-14 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 30.7 | 276.1 | – | – | 16384 | 2026-10-13 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 30.7 | 193.9 | – | – | 1024 | 2026-10-15 | community | contradiction | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 30.7 | 491.8 | – | – | 4096 | 2026-10-16 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 30.6 | 1117.0 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 30.6 | 1117.0 | – | – | 4096 | 2026-10-11 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 30.4 | 990.5 | – | – | 1024 | 2026-10-18 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 30.3 | 236.2 | – | – | 8192 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 30.3 | 236.2 | – | – | 8192 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 30.3 | 236.2 | – | – | 8192 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 30.3 | 236.2 | – | – | 8192 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 30.3 | 236.2 | – | – | 8192 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 30.3 | 236.2 | – | – | 8192 | 2026-03-09 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 30.3 | 493.4 | – | – | 4096 | 2026-10-16 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 30.1 | 217.6 | – | – | 16384 | 2026-10-07 | community | – | source |
| M5 Pro | Qwen3.5-9B | OptiQ-4bit | oMLX | 30.1 | 702.8 | – | – | 1024 | 2026-10-17 | community | – | source |
| M5 Pro | Qwen3.5-9B | 8bit | oMLX | 30.1 | 1251.0 | – | – | 16384 | 2026-10-17 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 30.0 | 138.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 30.0 | 138.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 30.0 | 138.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 30.0 | 138.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Pro (16 GB) | Qwen3.5-9B | 4bit | MLX | 30 | – | – | – | – | – | sourced | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.9 | 1059.0 | – | – | 1024 | 2026-10-10 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.9 | 1104.0 | – | – | 4096 | 2026-10-18 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.7 | 1022.0 | – | – | 1024 | 2026-10-07 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.7 | 1054.0 | – | – | 1024 | 2026-10-10 | community | contradiction | source |
| M4 | Qwen3.5-9B | oQ4 | oMLX | 29.6 | 224.0 | – | – | 1024 | 2026-10-08 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 29.5 | 167.4 | – | – | 16384 | 2026-10-16 | community | – | source |
| M5 Pro | Qwen3.5-9B | OptiQ-4bit | oMLX | 29.5 | 832.5 | – | – | 4096 | 2026-10-17 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.3 | 1119.0 | – | – | 4096 | 2026-10-10 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.2 | 1117.0 | – | – | 4096 | 2026-10-10 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.2 | 1095.0 | – | – | 16384 | 2026-10-18 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.1 | 1095.0 | – | – | 4096 | 2026-10-06 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 29.1 | 1121.0 | – | – | 4096 | 2026-10-07 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 28.9 | 285.5 | – | – | 4096 | 2026-10-13 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 28.9 | 475.4 | – | – | 4096 | 2026-10-13 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 28.8 | 1066.0 | – | – | 1024 | 2026-10-09 | community | contradiction | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 28.7 | 138.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 28.7 | 138.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 28.6 | 138.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 28.6 | 138.6 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 Pro | Qwen3.5-9B | 4bit | oMLX | 28.6 | 774.2 | – | – | 32768 | 2026-10-10 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 28.5 | 1083.0 | – | – | 16384 | 2026-10-11 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 28.5 | 1083.0 | – | – | 16384 | 2026-10-11 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 28.5 | 1101.0 | – | – | 4096 | 2026-10-17 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 28.4 | 273.3 | – | – | 1024 | 2026-10-13 | community | – | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 28.3 | 363.0 | – | – | 1024 | 2026-10-14 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 28.3 | 1046.0 | – | – | 1024 | 2026-10-17 | community | contradiction | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 28.2 | 363.3 | – | – | 1024 | 2026-10-14 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 28.0 | 1113.0 | – | – | 4096 | 2026-10-09 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 28.0 | 459.7 | – | – | 8192 | 2026-10-13 | community | – | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 27.9 | 351.7 | – | – | 1024 | 2026-10-14 | community | – | source |
| M2 Pro | Qwen3.5-9B | oQ6e | oMLX | 27.8 | 288.0 | – | – | 8192 | 2026-10-13 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 27.2 | 231.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 27.2 | 231.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 27.2 | 213.1 | – | – | 16384 | 2026-10-08 | community | – | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 27.0 | 293.2 | – | – | 4096 | 2026-10-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 27.0 | 685.5 | – | – | 1024 | 2026-10-15 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 26.8 | 833.7 | – | – | 1024 | 2026-10-17 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 26.7 | 829.4 | – | – | 4096 | 2026-10-15 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 26.6 | 1087.0 | – | – | 16384 | 2026-10-06 | community | – | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 26.6 | 373.2 | – | – | 8192 | 2026-10-14 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 26.5 | 798.6 | – | – | 1024 | 2026-10-07 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 26.4 | 837.7 | – | – | 1024 | 2026-10-06 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 26.4 | 837.7 | – | – | 1024 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 26.2 | 234.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 26.2 | 234.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 26.2 | 234.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 26.2 | 234.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 26.2 | 234.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 26.2 | 234.4 | – | – | 16384 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | MXFP4 | oMLX | 26.1 | 215.4 | – | – | 1024 | 2026-03-08 | community | – | source |
| M5 | Qwen3.5-9B | MXFP4 | oMLX | 26.1 | 215.4 | – | – | 1024 | 2026-03-08 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 26.1 | 875.5 | – | – | 4096 | 2026-10-17 | community | – | source |
| Arc A770 | Qwen3.5-9B | UD-Q4_K_XL | llama.cpp (Vulkan) | 26 | – | – | – | 131072 | 2026-09-27 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.9 | 235.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.9 | 235.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.9 | 708.1 | – | – | 1024 | 2026-10-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.9 | 836.6 | – | – | 1024 | 2026-10-13 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.8 | 763.7 | – | – | 4096 | 2026-10-08 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.8 | 838.4 | – | – | 4096 | 2026-10-13 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.7 | 862.3 | – | – | 4096 | 2026-10-08 | community | – | source |
| M5 Pro | Qwen3.5-9B | oQ4e | oMLX | 25.7 | 1029.0 | – | – | 1024 | 2026-10-15 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.6 | 916.5 | – | – | 4096 | 2026-10-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.5 | 885.3 | – | – | 4096 | 2026-10-06 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.5 | 885.3 | – | – | 4096 | 2026-10-06 | community | – | source |
| M5 | Qwen3.5-9B | MXFP4 | oMLX | 25.3 | 214.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M5 | Qwen3.5-9B | MXFP4 | oMLX | 25.3 | 214.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.3 | 231.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.3 | 231.9 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.1 | 825.4 | – | – | 16384 | 2026-10-17 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.0 | 232.5 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 25.0 | 232.5 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 Pro (64 GB) | Qwen3.5-9B | Q4_K - Medium | llama.cpp | 25 | – | – | 0.765 | – | – | sourced | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 24.6 | 872.4 | – | – | 4096 | 2026-10-07 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 24.6 | 873.9 | – | – | 8192 | 2026-10-17 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 24.5 | 180.4 | – | – | 1024 | 2026-10-12 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 24.5 | 956.3 | – | – | 32768 | 2026-10-14 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 24.5 | 506.7 | – | – | 1024 | 2026-10-14 | community | – | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 24.1 | 278.2 | – | – | 16384 | 2026-10-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 24.0 | 738.7 | – | – | 16384 | 2026-10-08 | community | – | source |
| M4 Pro | Qwen3.5-9B | 8bit | oMLX | 23.8 | 364.2 | – | – | 1024 | 2026-10-16 | community | contradiction | source |
| M5 Pro | Qwen3.5-9B | OptiQ-4bit | oMLX | 23.6 | 1346.0 | – | – | 1024 | 2026-10-16 | community | contradiction | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 23.6 | 750.3 | – | – | 32768 | 2026-10-17 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 23.4 | 843.1 | – | – | 16384 | 2026-10-08 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 23.3 | 560.4 | – | – | 4096 | 2026-10-14 | community | contradiction | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 23.2 | 200.8 | – | – | 4096 | 2026-10-12 | community | contradiction | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 22.9 | 859.4 | – | – | 4096 | 2026-10-06 | community | contradiction | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 22.8 | 838.6 | – | – | 16384 | 2026-10-06 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 22.2 | 217.5 | – | – | 131072 | 2026-10-12 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 22.2 | 596.7 | – | – | 199680 | 2026-10-13 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 22.1 | 176.4 | – | – | 1024 | 2026-10-12 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 22.0 | 235.4 | – | – | 1024 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.9 | 235.0 | – | – | 1024 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.8 | 234.8 | – | – | 1024 | 2026-10-14 | community | – | source |
| M4 | Qwen3.5-9B | MXFP4 | oMLX | 21.7 | 169.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.7 | 225.0 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | MXFP4 | oMLX | 21.7 | 169.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.7 | 225.0 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.6 | 234.3 | – | – | 1024 | 2026-10-14 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.6 | 242.1 | – | – | 4096 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.5 | 242.3 | – | – | 4096 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.5 | 234.4 | – | – | 1024 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.4 | 235.4 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.3 | 241.2 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.3 | 241.9 | – | – | 4096 | 2026-10-14 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.3 | 233.5 | – | – | 1024 | 2026-10-15 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.2 | 240.8 | – | – | 4096 | 2026-10-14 | community | – | source |
| Arc A770 | Qwen3.5-9B | UD-Q4_K_XL | llama.cpp (SYCL) | 21.19 | – | – | – | 131072 | 2026-09-27 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.1 | 240.2 | – | – | 8192 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.1 | 233.5 | – | – | 1024 | 2026-10-16 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 21.1 | 664.1 | – | – | 65536 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 225.4 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 225.4 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 222.5 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 222.5 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 203.7 | – | – | 1024 | 2026-10-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 203.7 | – | – | 1024 | 2026-10-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 21.0 | 180.9 | – | – | 4096 | 2026-10-12 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 21.0 | 232.7 | – | – | 1024 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 196.5 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 196.5 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 241.2 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 241.3 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 242.0 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 235.4 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.9 | 241.5 | – | – | 4096 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | oQ4 | oMLX | 20.8 | 234.2 | – | – | 1024 | 2026-10-08 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.8 | 232.2 | – | – | 1024 | 2026-10-17 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 20.8 | 849.3 | – | – | 65536 | 2026-10-18 | community | – | source |
| M4 | Qwen3.5-9B | MXFP4 | oMLX | 20.7 | 169.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 224.8 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | MXFP4 | oMLX | 20.7 | 169.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 224.8 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 222.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 222.2 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 222.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 229.9 | – | – | 1024 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 237.3 | – | – | 4096 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 229.9 | – | – | 1024 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 237.3 | – | – | 4096 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 274.6 | – | – | 4096 | 2026-10-15 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.7 | 227.0 | – | – | 1024 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.6 | 222.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.5 | 241.9 | – | – | 4096 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 226.4 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 234.6 | – | – | 8192 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 225.6 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 241.3 | – | – | 4096 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 199.6 | – | – | 1024 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 240.1 | – | – | 4096 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.4 | 241.1 | – | – | 4096 | 2026-10-17 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 20.3 | 224.8 | – | – | 32768 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 20.3 | 224.8 | – | – | 32768 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 20.3 | 224.8 | – | – | 32768 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.3 | 196.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 20.3 | 224.8 | – | – | 32768 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.3 | 196.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 20.3 | 224.8 | – | – | 32768 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 20.3 | 224.8 | – | – | 32768 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.3 | 234.9 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.3 | 234.9 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.3 | 239.5 | – | – | 8192 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.2 | 224.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.2 | 224.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | oQ4 | oMLX | 20.2 | 234.3 | – | – | 4096 | 2026-10-08 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.2 | 235.8 | – | – | 4096 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 223.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 222.7 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 223.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 222.7 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 215.2 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 215.2 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 237.5 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 237.5 | – | – | 4096 | 2026-10-11 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 795.3 | – | – | 65536 | 2026-10-14 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 204.5 | – | – | 4096 | 2026-10-16 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.1 | 240.0 | – | – | 8192 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 20.0 | 234.6 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.9 | 219.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.9 | 219.1 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.8 | 218.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.8 | 218.1 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.8 | 224.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.8 | 224.2 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 19.7 | 191.4 | – | – | 4096 | 2026-10-12 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.6 | 223.7 | – | – | 1024 | 2026-03-08 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.6 | 223.7 | – | – | 1024 | 2026-03-08 | community | contradiction | source |
| M1 Pro | Qwen3.5-9B | 8bit | oMLX | 19.5 | 194.9 | – | – | 1024 | 2026-10-07 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 19.5 | 233.0 | – | – | 1024 | 2026-10-15 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.4 | 239.4 | – | – | 4096 | 2026-10-06 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.4 | 239.4 | – | – | 4096 | 2026-10-06 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.4 | 234.5 | – | – | 4096 | 2026-10-13 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.2 | 165.4 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.2 | 165.4 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.2 | 230.1 | – | – | 16384 | 2026-10-17 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 19.1 | 147.0 | – | – | 1024 | 2026-10-12 | community | contradiction | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 19.1 | 194.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.0 | 225.0 | – | – | 1024 | 2026-10-06 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.0 | 225.0 | – | – | 1024 | 2026-10-06 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.0 | 155.3 | – | – | 1024 | 2026-03-09 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.0 | 155.3 | – | – | 1024 | 2026-03-09 | community | contradiction | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 19.0 | 196.0 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 19.0 | 239.6 | – | – | 4096 | 2026-10-15 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 19.0 | 230.4 | – | – | 16384 | 2026-10-16 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 18.9 | 195.7 | – | – | 1024 | 2026-10-14 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.8 | 110.4 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.8 | 110.4 | – | – | 1024 | 2026-10-11 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.8 | 110.4 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 18.8 | 231.0 | – | – | 16384 | 2026-10-17 | community | – | source |
| M1 Pro | Qwen3.5-9B | 8bit | oMLX | 18.7 | 199.9 | – | – | 4096 | 2026-10-07 | community | – | source |
| M5 | Qwen3.5-9B | OptiQ-4bit | oMLX | 18.7 | 749.0 | – | – | 4096 | 2026-10-08 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 18.7 | 202.3 | – | – | 4096 | 2026-10-14 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 18.6 | 228.2 | – | – | 16384 | 2026-10-08 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.6 | 105.8 | – | – | 4096 | 2026-10-11 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.6 | 105.8 | – | – | 4096 | 2026-10-11 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.6 | 105.8 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 231.2 | – | – | 16384 | 2026-10-08 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 86.4 | – | – | 4096 | 2026-10-12 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 193.7 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 232.9 | – | – | 1024 | 2026-10-14 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 239.4 | – | – | 4096 | 2026-10-14 | community | contradiction | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 201.8 | – | – | 4096 | 2026-10-17 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.5 | 109.9 | – | – | 1024 | 2026-10-17 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 18.3 | 226.9 | – | – | 16384 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 18.3 | 233.2 | – | – | 4096 | 2026-10-15 | community | contradiction | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 18.3 | 105.5 | – | – | 4096 | 2026-10-17 | community | – | source |
| M1 Pro | Qwen3.5-9B | 8bit | oMLX | 18.2 | 185.1 | – | – | 1024 | 2026-10-14 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 17.8 | 155.6 | – | – | 4096 | 2026-03-09 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 17.8 | 155.6 | – | – | 4096 | 2026-03-09 | community | contradiction | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 17.8 | 101.5 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 17.8 | 101.5 | – | – | 8192 | 2026-10-11 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 17.8 | 101.5 | – | – | 8192 | 2026-10-11 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 17.7 | 174.7 | – | – | 1024 | 2026-03-09 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 17.7 | 174.7 | – | – | 1024 | 2026-03-09 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 17.6 | 157.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 17.6 | 157.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 17.6 | 104.7 | – | – | 4096 | 2026-10-14 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 17.6 | 540.3 | – | – | 131072 | 2026-10-17 | community | – | source |
| M5 | Qwen3.5-9B | OptiQ-4bit | oMLX | 17.5 | 734.2 | – | – | 16384 | 2026-10-08 | community | – | source |
| M1 Pro | Qwen3.5-9B | 8bit | oMLX | 17.5 | 247.9 | – | – | 4096 | 2026-10-14 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 17.4 | 85.4 | – | – | 8192 | 2026-10-12 | community | – | source |
| M6 | Qwen3.5-9B | 8bit | oMLX | 17.4 | 892.6 | – | – | 1024 | 2026-10-13 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 17.3 | 131.1 | – | – | 1024 | 2026-03-09 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 17.3 | 131.1 | – | – | 1024 | 2026-03-09 | community | contradiction | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 17.3 | 199.6 | – | – | 8192 | 2026-10-17 | community | – | source |
| M3 Pro | Qwen3.5-9B | 4bit | oMLX | 17.2 | 235.9 | – | – | 1024 | 2026-10-10 | community | contradiction | source |
| M6 | Qwen3.5-9B | 8bit | oMLX | 17.1 | 973.6 | – | – | 4096 | 2026-10-13 | community | – | source |
| M3 | Qwen3.5-9B | 4bit | oMLX | 17.0 | 195.7 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 17.0 | 228.7 | – | – | 16384 | 2026-10-08 | community | contradiction | source |
| M4 | Qwen3.5-9B | 5bit | oMLX | 17.0 | 160.1 | – | – | 1024 | 2026-10-15 | community | – | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 16.9 | 130.7 | – | – | 4096 | 2026-03-09 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | 4bit | oMLX | 16.9 | 130.7 | – | – | 4096 | 2026-03-09 | community | contradiction | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 16.9 | 606.3 | – | – | 8192 | 2026-10-12 | community | contradiction | source |
| M4 | Qwen3.5-9B | 5bit | oMLX | 16.9 | 242.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M4 | Qwen3.5-9B | oQ4e | oMLX | 16.8 | 212.2 | – | – | 16384 | 2026-10-15 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 16.7 | 167.8 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 16.7 | 167.8 | – | – | 4096 | 2026-10-11 | community | – | source |
| M4 Pro | Qwen3.5-9B | 4bit | oMLX | 16.7 | 176.2 | – | – | 199680 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 16.2 | 208.3 | – | – | 4096 | 2026-10-06 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 16.2 | 208.3 | – | – | 4096 | 2026-10-06 | community | contradiction | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 16.0 | 588.3 | – | – | 16384 | 2026-10-12 | community | contradiction | source |
| M4 | Qwen3.5-9B | OptiQ-4bit | oMLX | 15.9 | 152.2 | – | – | 1024 | 2026-10-15 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.8 | 217.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.7 | 218.0 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.7 | 216.9 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.7 | 218.0 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.7 | 217.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.7 | 217.6 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | Qwen3.5-9B | 4bit | oMLX | 15.7 | 533.3 | – | – | 4096 | 2026-10-14 | community | contradiction outlier | source |
| M4 | Qwen3.5-9B | OptiQ-4bit | oMLX | 15.6 | 198.1 | – | – | 4096 | 2026-10-15 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.3 | 218.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M4 | Qwen3.5-9B | 6bit | oMLX | 15.3 | 218.9 | – | – | 4096 | 2026-03-08 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 15.3 | 478.1 | – | – | 4096 | 2026-10-12 | community | contradiction | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 15.0 | 447.6 | – | – | 199680 | 2026-10-17 | community | – | source |
| M5 | Qwen3.5-9B | MXFP4 | oMLX | 14.8 | 340.3 | – | – | 4096 | 2026-10-07 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 14.8 | 643.6 | – | – | 131072 | 2026-10-18 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 14.6 | 43.8 | – | – | 16384 | 2026-10-12 | community | – | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 14.3 | 201.7 | – | – | 16384 | 2026-10-06 | community | contradiction | source |
| M4 | Qwen3.5-9B | 4bit | oMLX | 14.3 | 201.7 | – | – | 16384 | 2026-10-06 | community | contradiction | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 14.3 | 708.5 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | OptiQ | oMLX | 14.2 | 224.2 | – | – | 1024 | 2026-10-13 | community | – | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 14.1 | 707.6 | – | – | 8192 | 2026-10-13 | community | – | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 14.0 | 590.9 | – | – | 131072 | 2026-10-14 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 13.9 | 205.0 | – | – | 65536 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 13.9 | 205.0 | – | – | 65536 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 13.9 | 205.0 | – | – | 65536 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 13.9 | 205.0 | – | – | 65536 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 13.9 | 205.0 | – | – | 65536 | 2026-03-09 | community | – | source |
| M1 Max | Qwen3.5-9B | 8bit | oMLX | 13.9 | 205.0 | – | – | 65536 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 13.9 | 695.7 | – | – | 4096 | 2026-10-11 | community | – | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 13.9 | 695.7 | – | – | 4096 | 2026-10-11 | community | – | source |
| M5 | Qwen3.5-9B | MXFP4 | oMLX | 13.7 | 327.8 | – | – | 16384 | 2026-10-07 | community | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.6 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.6 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.6 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 13.6 | 664.1 | – | – | 16384 | 2026-10-13 | community | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.5 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.3 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.2 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.2 | – | – | – | 128 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.2 | – | – | – | 8192 | 2026-10-13 | sourced | – | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 13.2 | 690.9 | – | – | 16384 | 2026-10-11 | community | – | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 13.2 | 690.9 | – | – | 16384 | 2026-10-11 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 13.2 | 462.2 | – | – | 16384 | 2026-10-12 | community | contradiction | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.1 | – | – | – | 8192 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.0 | – | – | – | 8192 | 2026-10-13 | sourced | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 13.0 | – | – | – | 8192 | 2026-10-13 | sourced | – | source |
| M1 | Qwen3.5-9B | 4bit | oMLX | 12.6 | 77.8 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 | Qwen3.5-9B | 4bit | oMLX | 12.6 | 77.8 | – | – | 1024 | 2026-03-09 | community | – | source |
| M5 | Qwen3.5-9B | 4bit | oMLX | 12.6 | 471.4 | – | – | 32768 | 2026-10-12 | community | contradiction | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 12.4 | 416.0 | – | – | 4096 | 2026-10-18 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.3 | 127.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.3 | 127.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.3 | 127.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.3 | 127.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.3 | 127.6 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 | Qwen3.5-9B | 4bit | oMLX | 12.1 | 77.5 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 | Qwen3.5-9B | 4bit | oMLX | 12.1 | 77.5 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.0 | 132.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.0 | 132.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.0 | 132.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.0 | 132.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 12.0 | 132.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 11.9 | – | – | – | 128 | 2026-10-13 | sourced | contradiction | source |
| M5 | Qwen3.5-9B | 8bit | oMLX | 11.7 | 441.8 | – | – | 16384 | 2026-10-18 | community | contradiction | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 11.5 | 130.7 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 11.5 | 130.7 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 11.5 | 130.7 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 11.5 | 130.7 | – | – | 8192 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 11.5 | 130.7 | – | – | 8192 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 8bit | oMLX | 11.0 | 222.9 | – | – | 1024 | 2026-10-10 | community | – | source |
| M4 | Qwen3.5-9B | 8bit | oMLX | 10.7 | 217.4 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 | Qwen3.5-9B | 8bit | oMLX | 10.7 | 220.4 | – | – | 4096 | 2026-10-13 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 10.6 | 127.5 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 10.6 | 127.5 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 10.6 | 127.5 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 10.6 | 127.5 | – | – | 16384 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 10.6 | 127.5 | – | – | 16384 | 2026-03-09 | community | – | source |
| M4 | Qwen3.5-9B | 8bit | oMLX | 10.4 | 218.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M2 | Qwen3.5-9B | 4bit | oMLX | 10.2 | 77.9 | – | – | 16384 | 2026-10-14 | community | contradiction | source |
| M6 | Qwen3.5-9B | 4bit | oMLX | 10.1 | 473.8 | – | – | 199680 | 2026-10-14 | community | contradiction | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 9.1 | – | – | – | 8192 | 2026-10-13 | sourced | contradiction | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 9.0 | – | – | – | 8192 | 2026-10-13 | sourced | contradiction | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 8.8 | 241.1 | – | – | 16384 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 8.8 | 241.1 | – | – | 16384 | 2026-10-06 | community | – | source |
| M1 Max | Qwen3.5-9B | 4bit | oMLX | 8.8 | 241.1 | – | – | 16384 | 2026-10-06 | community | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 8.6 | – | – | – | 128 | 2026-10-13 | sourced | contradiction | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 8.6 | 90.4 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 8.6 | 90.4 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 8.6 | 90.4 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 8.6 | 90.4 | – | – | 32768 | 2026-03-09 | community | – | source |
| M2 Pro | Qwen3.5-9B | 8bit | oMLX | 8.6 | 90.4 | – | – | 32768 | 2026-03-09 | community | – | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 8.1 | – | – | – | 8192 | 2026-10-13 | sourced | – | source |
| M1 Pro | Qwen3.5-9B | 4bit | oMLX | 8.1 | 185.5 | – | – | 4096 | 2026-10-15 | community | contradiction outlier | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 6.5 | – | – | – | 8192 | 2026-10-13 | sourced | contradiction | source |
| gfx1151 | Qwen3.5-9B | unstated | vLLM | 6.4 | – | – | – | 8192 | 2026-10-13 | sourced | contradiction | source |
| M1 | Qwen3.5-9B | 4bit | oMLX | 6.1 | 39.4 | – | – | 1024 | 2026-10-13 | community | contradiction | source |
| M1 | Qwen3.5-9B | 4bit | oMLX | 5.7 | 39.0 | – | – | 4096 | 2026-10-13 | community | contradiction | source |
| M4 (16 GB) | Qwen3.5-9B | Q4_0 | llama.cpp | 4.0648 | – | – | 0.5057 | – | – | sourced | – | source |
| M4 (16 GB) | Qwen3.5-9B | Q4_K - Small | llama.cpp | 3.142 | – | – | 0.8445 | – | – | sourced | – | source |
| M4 (16 GB) | Qwen3.5-9B | Q6_K | llama.cpp | 2.1849 | – | – | 1.5031 | – | – | sourced | – | source |
| RTX 4090 | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | – | 9343.1 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | – | 9515.0 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | – | 9472.7 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | Qwen3.5-9B | Q4_K_M | llama.cpp (CUDA) | – | 9619.7 | – | – | – | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | – | 1029 | – | – | – | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | – | 1028 | – | – | – | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | – | 607 | – | – | – | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | – | 1005 | – | – | – | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | – | 767 | – | – | – | 2026-10 | community | – | source |
| Arc Pro B50 | Qwen3.5-9B | Q4_0 | llama.cpp (SYCL) | – | 1150 | – | – | – | 2026-10 | community | – | source |