← Back to community benchmarks
gpt-oss-120b-MXFP4-Q8
Performance
1k
tokens
613.2
PP tok/s
68.4
TG tok/s
1670
TTFT (ms)
59.2
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 15.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 613.2 | 68.4 | 59.2 GB | current |
| 4k | 685.5 | 64.1 | 59.3 GB | view |
| 8k | 687.8 | 57.8 | 59.5 GB | view |
| 16k | 663.6 | 47.5 | 59.8 GB | view |
| 32k | 600.7 | 35.2 | 60.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 68.4 | 1.00× |
| 2× | 105.7 | 1.55× |
| 4× | 145.9 | 2.13× |
| 8× | 187.1 | 2.74× |