GLM-4.7-Flash
50 records across 14 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M5 Max | GLM-4.7-Flash | 4bit | oMLX | 93.2 | 694.2 | – | – | 1024 | 2026-10-13 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 4bit | oMLX | 75.5 | 1823.0 | – | – | 1024 | 2026-10-16 | community | – | source |
| M5 Max | GLM-4.7-Flash | 8bit | oMLX | 71.9 | 824.5 | – | – | 1024 | 2026-10-15 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 4bit | oMLX | 66.8 | 572.0 | – | – | 1024 | 2026-10-09 | community | – | source |
| M4 Max | GLM-4.7-Flash | 6bit | oMLX | 64.1 | 1087.0 | – | – | 4096 | 2026-10-08 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 4bit | oMLX | 61.8 | 1604.0 | – | – | 4096 | 2026-10-16 | community | – | source |
| M3 Max | GLM-4.7-Flash | 6bit | oMLX | 61.5 | 987.2 | – | – | 1024 | 2026-10-12 | community | – | source |
| M5 Max | GLM-4.7-Flash | 8bit | oMLX | 61.3 | 1937.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 4bit | oMLX | 60.9 | 1494.0 | – | – | 4096 | 2026-10-18 | community | – | source |
| M4 Max | GLM-4.7-Flash | 8bit | oMLX | 60.2 | 1160.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 Max | GLM-4.7-Flash | 8bit | oMLX | 60.2 | 1160.0 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 Max | GLM-4.7-Flash | 8bit | oMLX | 59.8 | 1213.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M3 Ultra (256 GB) | GLM-4.7-Flash | 8bit | MLX | 58 | – | – | – | – | – | sourced | – | source |
| M4 Max | GLM-4.7-Flash | 8bit | oMLX | 56.6 | 1114.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 4bit | oMLX | 55.7 | 1101.0 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 Max | GLM-4.7-Flash | 8bit | oMLX | 55.3 | 1091.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 Max | GLM-4.7-Flash | 8bit | oMLX | 55.3 | 1091.0 | – | – | 4096 | 2026-03-09 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 8bit | oMLX | 52.3 | 1319.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 8bit | oMLX | 52.2 | 1002.0 | – | – | 1024 | 2026-10-14 | community | – | source |
| M4 Pro | GLM-4.7-Flash | 6bit | oMLX | 49.8 | 608.4 | – | – | 1024 | 2026-03-09 | community | – | source |
| M4 Pro | GLM-4.7-Flash | 6bit | oMLX | 49.8 | 608.4 | – | – | 1024 | 2026-03-09 | community | – | source |
| M1 Ultra | GLM-4.7-Flash | 4bit | oMLX | 49.1 | 674.7 | – | – | 1024 | 2026-10-16 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 4bit | oMLX | 48.8 | 753.5 | – | – | 16384 | 2026-10-18 | community | – | source |
| M1 Ultra | GLM-4.7-Flash | 4bit | oMLX | 45.2 | 742.2 | – | – | 4096 | 2026-10-16 | community | – | source |
| M3 Max | GLM-4.7-Flash | 6bit | oMLX | 43.5 | 621.8 | – | – | 1024 | 2026-10-14 | community | contradiction | source |
| M5 Pro | GLM-4.7-Flash | 8bit | oMLX | 43.3 | 1258.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M5 Pro | GLM-4.7-Flash | 8bit | oMLX | 43.3 | 1301.0 | – | – | 4096 | 2026-10-14 | community | – | source |
| M4 Pro | GLM-4.7-Flash | 6bit | oMLX | 42.5 | 546.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M4 Pro | GLM-4.7-Flash | 6bit | oMLX | 42.5 | 546.4 | – | – | 4096 | 2026-03-09 | community | – | source |
| M1 Ultra | GLM-4.7-Flash | 4bit | oMLX | 39.2 | 634.5 | – | – | 8192 | 2026-10-16 | community | – | source |
| M3 Ultra | GLM-4.7-Flash | oQ4 | oMLX | 37.5 | 520.2 | – | – | 32768 | 2026-10-08 | community | – | source |
| M1 Max (64 GB) | GLM-4.7-Flash | Q4_K_XL | llama.cpp | 36.8 | 99.4 | – | – | 4096 | – | sourced | – | source |
| M5 | GLM-4.7-Flash | oQ5 | oMLX | 36.7 | 833.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M1 Pro | GLM-4.7-Flash | 4bit | oMLX | 33.7 | 292.6 | – | – | 1024 | 2026-10-15 | community | – | source |
| M1 Pro | GLM-4.7-Flash | 4bit | oMLX | 33.6 | 295.8 | – | – | 1024 | 2026-10-15 | community | – | source |
| M1 Ultra | GLM-4.7-Flash | 4bit | oMLX | 32.7 | 472.0 | – | – | 16384 | 2026-10-16 | community | – | source |
| M5 | GLM-4.7-Flash | oQ5 | oMLX | 31.4 | 613.5 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 | GLM-4.7-Flash | 4bit | oMLX | 29.9 | 286.7 | – | – | 4096 | 2026-10-14 | community | – | source |
| M1 Pro | GLM-4.7-Flash | 4bit | oMLX | 28.6 | 260.2 | – | – | 4096 | 2026-10-15 | community | – | source |
| M1 Pro | GLM-4.7-Flash | 4bit | oMLX | 27.9 | 260.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M3 Ultra | GLM-4.7-Flash | oQ4 | oMLX | 21.3 | 288.8 | – | – | 65536 | 2026-10-08 | community | – | source |
| Radeon 780M | GLM-4.7-Flash | Q4_K_M | llama.cpp (Vulkan) | 12.91 | – | – | – | 32768 | 2026-10-05 | community | – | source |
| M3 Max | GLM-4.7-Flash | 6bit | oMLX | 11.9 | 120.9 | – | – | 65536 | 2026-10-14 | community | – | source |
| Radeon 780M | GLM-4.7-Flash | Q4_K_M | llama.cpp (Vulkan) | 8.83 | – | – | – | 32768 | 2026-10-05 | community | – | source |
| RTX 4090 | GLM-4.7-Flash | Q4_K | llama.cpp (CUDA) | – | 6750.5 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | GLM-4.7-Flash | Q4_K | llama.cpp (CUDA) | – | 6843.1 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | GLM-4.7-Flash | Q4_K | llama.cpp (CUDA) | – | 5732.7 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | GLM-4.7-Flash | Q4_K | llama.cpp (CUDA) | – | 5794.6 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | GLM-4.7-Flash | Q4_K | llama.cpp (CUDA) | – | 8437.7 | – | – | – | 2026-10 | community | – | source |
| RTX 4090 | GLM-4.7-Flash | Q4_K | llama.cpp (CUDA) | – | 8587.5 | – | – | – | 2026-10 | community | – | source |