LFM2.5-8B-A1B
66 records across 11 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M4 Max | LFM2.5-8B-A1B | MXFP4 | oMLX | 272.8 | 3549.0 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 Max | LFM2.5-8B-A1B | MXFP4 | oMLX | 264.3 | 4071.0 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 Max | LFM2.5-8B-A1B | MXFP4 | oMLX | 256.2 | 3968.0 | – | – | 8192 | 2026-10-08 | community | – | source |
| RTX 4080 SUPER | LFM2.5-8B-A1B | IQ1_S | llama.cpp (CUDA) | 241.8 | – | – | – | 8192 | 2026-10-04 | community | – | source |
| RTX 4080 SUPER | LFM2.5-8B-A1B | IQ2_XXS | llama.cpp (CUDA) | 224.1 | – | – | – | 8192 | 2026-10-04 | community | – | source |
| RTX 4080 SUPER | LFM2.5-8B-A1B | IQ2_M | llama.cpp (CUDA) | 208.5 | – | – | – | 8192 | 2026-10-04 | community | – | source |
| RTX 4080 SUPER | LFM2.5-8B-A1B | IQ3_XXS | llama.cpp (CUDA) | 196.2 | – | – | – | 8192 | 2026-10-04 | community | – | source |
| M5 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 179.7 | 6365.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M5 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 179.3 | 6307.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M5 Max | LFM2.5-8B-A1B | 8bit | oMLX | 174.9 | 12888.0 | – | – | 4096 | 2026-10-07 | community | – | source |
| M5 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 173.1 | 6212.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M2 Max | LFM2.5-8B-A1B | 4bit | oMLX | 173.0 | 1897.0 | – | – | 1024 | 2026-10-18 | community | – | source |
| RTX 4080 SUPER | LFM2.5-8B-A1B | Q4_K_M | llama.cpp (CUDA) | 172.4 | – | – | – | 8192 | 2026-10-04 | community | – | source |
| M5 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 165.7 | 5676.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M2 Max | LFM2.5-8B-A1B | 4bit | oMLX | 163.3 | 2088.0 | – | – | 4096 | 2026-10-18 | community | – | source |
| M5 Max | LFM2.5-8B-A1B | 8bit | oMLX | 152.6 | 12166.0 | – | – | 16384 | 2026-10-07 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 146.4 | 1614.0 | – | – | 1024 | 2026-10-11 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 146.3 | 1633.0 | – | – | 1024 | 2026-10-11 | community | – | source |
| M5 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 142.9 | 5520.0 | – | – | 1024 | 2026-10-09 | community | contradiction | source |
| RTX 4080 SUPER | LFM2.5-8B-A1B | unspecified | llama.cpp (CUDA) | 118.5 | – | – | – | 8192 | 2026-10-04 | community | – | source |
| M5 Max | LFM2.5-8B-A1B | 8bit | oMLX | 108.5 | 9844.0 | – | – | 65536 | 2026-10-07 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 107.0 | 1369.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 107.0 | 1369.0 | – | – | 1024 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 106.1 | 1619.0 | – | – | 4096 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 106.1 | 1619.0 | – | – | 4096 | 2026-10-13 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 104.5 | 3428.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 104.5 | 3428.0 | – | – | 4096 | 2026-10-06 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 103.1 | 1582.0 | – | – | 8192 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 103.1 | 1582.0 | – | – | 8192 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 94.5 | 1495.0 | – | – | 16384 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 94.5 | 1495.0 | – | – | 16384 | 2026-10-13 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 93.7 | 826.2 | – | – | 4096 | 2026-10-11 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 91.9 | 3179.0 | – | – | 16384 | 2026-10-06 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 91.9 | 3179.0 | – | – | 16384 | 2026-10-06 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 87.8 | 777.2 | – | – | 16384 | 2026-10-17 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 86.8 | 1545.0 | – | – | 1024 | 2026-10-13 | community | contradiction | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 85.7 | 1643.0 | – | – | 4096 | 2026-10-13 | community | contradiction | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 81.1 | 1338.0 | – | – | 32768 | 2026-10-13 | community | – | source |
| M4 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 81.1 | 1338.0 | – | – | 32768 | 2026-10-13 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 80.0 | 2898.0 | – | – | 32768 | 2026-10-06 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 80.0 | 2898.0 | – | – | 32768 | 2026-10-06 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 77.2 | 620.5 | – | – | 1024 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 77.2 | 620.5 | – | – | 1024 | 2026-10-17 | community | – | source |
| M4 | LFM2.5-8B-A1B | 4bit | oMLX | 76.0 | 965.1 | – | – | 1024 | 2026-10-13 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 74.6 | 685.6 | – | – | 4096 | 2026-10-17 | community | – | source |
| M3 Ultra | LFM2.5-8B-A1B | unknown | oMLX | 74.1 | 1262.0 | – | – | 65536 | 2026-10-13 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 71.2 | 669.8 | – | – | 1024 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 69.7 | 682.7 | – | – | 8192 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 69.7 | 682.7 | – | – | 8192 | 2026-10-17 | community | – | source |
| M3 | LFM2.5-8B-A1B | 4bit | oMLX | 68.7 | 805.5 | – | – | 4096 | 2026-10-16 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 68.5 | 736.9 | – | – | 4096 | 2026-10-17 | community | – | source |
| M4 | LFM2.5-8B-A1B | 4bit | oMLX | 66.9 | 1006.0 | – | – | 4096 | 2026-10-13 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 65.6 | 731.7 | – | – | 8192 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 64.0 | 643.1 | – | – | 16384 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 60.7 | 692.2 | – | – | 16384 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 53.6 | 596.7 | – | – | 32768 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 53.6 | 596.7 | – | – | 32768 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 52.1 | 637.0 | – | – | 32768 | 2026-10-17 | community | – | source |
| M3 | LFM2.5-8B-A1B | 4bit | oMLX | 51.0 | 617.5 | – | – | 32768 | 2026-10-16 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 42.2 | 507.7 | – | – | 65536 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 42.2 | 507.7 | – | – | 65536 | 2026-10-17 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 8bit | oMLX | 41.0 | 539.1 | – | – | 65536 | 2026-10-17 | community | – | source |
| M3 Ultra | LFM2.5-8B-A1B | unknown | oMLX | 37.5 | 2866.0 | – | – | 32768 | 2026-10-13 | community | – | source |
| M1 Pro | LFM2.5-8B-A1B | 4bit | oMLX | 32.9 | 463.1 | – | – | 131072 | 2026-10-17 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 19.5 | 1175.0 | – | – | 131072 | 2026-10-06 | community | – | source |
| M5 | LFM2.5-8B-A1B | 4bit | oMLX | 19.5 | 1175.0 | – | – | 131072 | 2026-10-06 | community | – | source |