gpt-oss 120B
96 records across 10 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| H100 | gpt-oss 120B | as-published | vLLM | 119092 | – | – | – | – | 2026-09-25 | community | – | source |
| H100 | gpt-oss 120B | as-published | vLLM | 102876 | – | – | – | – | 2026-09-25 | community | – | source |
| H200 | gpt-oss 120B 120B MoE | as-published | vLLM | 275.84 | – | – | – | – | 2026-03 | community | – | source |
| H200 | gpt-oss 120B 120B MoE | as-published | vLLM | 275.84 | – | – | – | – | 2026-03 | community | – | source |
| H200 | gpt-oss 120B 120B MoE | as-published | vLLM | 231.22 | – | – | – | – | 2026-03 | community | – | source |
| RTX Pro 6000 Blackwell | gpt-oss 120B 120B | MXFP4 | llama.cpp | 210.1 | 4578.6 | – | – | 4096 | – | sourced | – | source |
| RTX Pro 6000 Blackwell | gpt-oss 120B 120B | MXFP4 | llama.cpp | 182.2 | 4060.7 | – | – | 16384 | – | sourced | – | source |
| RTX Pro 6000 Blackwell | gpt-oss 120B 120B | MXFP4 | llama.cpp | 161.5 | 3368.3 | – | – | 32768 | – | sourced | – | source |
| RTX Pro 6000 Blackwell | gpt-oss 120B 120B | MXFP4 | llama.cpp | 133.1 | 2360.3 | – | – | 65536 | – | sourced | – | source |
| RTX Pro 6000 Blackwell | gpt-oss 120B 120B | MXFP4 | llama.cpp | 99.8 | 1289.8 | – | – | 131072 | – | sourced | – | source |
| M5 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 97.2 | 858.1 | – | – | 1024 | 2026-10-08 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4-Q8 | oMLX | 93.0 | 885.3 | – | – | 1024 | 2026-10-15 | community | – | source |
| M5 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 90.2 | 3148.0 | – | – | 8192 | 2026-10-08 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4-Q8 | oMLX | 89.4 | 2061.0 | – | – | 4096 | 2026-10-15 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4 | oMLX | 81.2 | 727.7 | – | – | 1024 | 2026-10-11 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4 | oMLX | 81.2 | 727.7 | – | – | 1024 | 2026-10-11 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4 | oMLX | 78.5 | 2110.0 | – | – | 4096 | 2026-10-11 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4 | oMLX | 78.5 | 2110.0 | – | – | 4096 | 2026-10-11 | community | – | source |
| M5 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 75.4 | 2941.0 | – | – | 32768 | 2026-10-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 74.3 | 876.2 | – | – | 1024 | 2026-10-09 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 72.5 | 1072.0 | – | – | 4096 | 2026-10-09 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 69.6 | 1068.0 | – | – | 8192 | 2026-10-09 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 68.4 | 613.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 68.4 | 613.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 68.4 | 613.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 68.4 | 613.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 68.4 | 613.2 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 67.7 | 314.1 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 67.7 | 314.1 | – | – | 1024 | 2026-03-08 | community | – | source |
| Arc Pro B70 | gpt-oss 120B | MXFP4 | llama.cpp (SYCL) | 67.14 | – | – | – | – | 2026-10-01 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 65.5 | 1011.0 | – | – | 16384 | 2026-10-09 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 64.1 | 685.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 64.1 | 685.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 64.1 | 685.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 64.1 | 685.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 64.1 | 685.5 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 60.7 | 569.3 | – | – | 1024 | 2026-03-08 | community | contradiction | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 60.7 | 569.3 | – | – | 1024 | 2026-03-08 | community | contradiction | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 60.1 | 363.7 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 60.1 | 363.7 | – | – | 4096 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 59.0 | 915.4 | – | – | 32768 | 2026-10-09 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 57.8 | 687.8 | – | – | 8192 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 57.8 | 687.8 | – | – | 8192 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 57.8 | 687.8 | – | – | 8192 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 57.8 | 687.8 | – | – | 8192 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 57.8 | 687.8 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 54.9 | 187.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 54.9 | 187.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 54.9 | 187.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 54.9 | 187.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 54.9 | 187.7 | – | – | 1024 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 54.2 | 631.9 | – | – | 4096 | 2026-03-08 | community | contradiction | source |
| M3 Ultra | gpt-oss 120B | 8bit | oMLX | 54.2 | 631.9 | – | – | 4096 | 2026-03-08 | community | contradiction | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 47.5 | 663.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 47.5 | 663.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 47.5 | 663.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 47.5 | 663.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 47.5 | 663.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 44.7 | 182.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 44.7 | 182.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 44.7 | 182.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 44.7 | 182.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 44.7 | 182.0 | – | – | 4096 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 43.7 | 365.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 43.7 | 365.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 38.4 | 361.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 38.4 | 361.6 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 37.1 | 183.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 37.1 | 183.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 37.1 | 183.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 37.1 | 183.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 37.1 | 183.9 | – | – | 8192 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 35.2 | 600.7 | – | – | 32768 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 35.2 | 600.7 | – | – | 32768 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 35.2 | 600.7 | – | – | 32768 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 35.2 | 600.7 | – | – | 32768 | 2026-03-08 | community | – | source |
| M3 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 35.2 | 600.7 | – | – | 32768 | 2026-03-08 | community | – | source |
| M5 Max | gpt-oss 120B | MXFP4 | oMLX | 34.1 | 1318.0 | – | – | 131072 | 2026-10-11 | community | – | source |
| Arc Pro B70 | gpt-oss 120B | MXFP4 | llama.cpp (SYCL) | 33.27 | – | – | – | – | 2026-10-01 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 25.7 | 337.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 25.7 | 337.1 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 22.7 | 177.1 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 22.7 | 177.1 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 22.7 | 177.1 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 22.7 | 177.1 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 22.7 | 177.1 | – | – | 16384 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 15.4 | 294.5 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Ultra | gpt-oss 120B | MXFP4-Q8 | oMLX | 15.4 | 294.5 | – | – | 65536 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 14.6 | 165.3 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 14.6 | 165.3 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 14.6 | 165.3 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 14.6 | 165.3 | – | – | 32768 | 2026-03-08 | community | – | source |
| M2 Max | gpt-oss 120B | 4bit | oMLX | 14.6 | 165.3 | – | – | 32768 | 2026-03-08 | community | – | source |
| M4 Ultra (192 GB) | gpt-oss 120B | Q4_K - Medium | MLX | 13 | – | – | 4.2 | – | – | sourced | – | source |
| Arc Pro B70 | gpt-oss 120B | MXFP4 | llama.cpp (SYCL) | – | 594.5 | – | – | – | 2026-10-01 | community | – | source |
| Arc Pro B70 | gpt-oss 120B | MXFP4 | llama.cpp (SYCL) | – | 600.1 | – | – | – | 2026-10-01 | community | – | source |