← Back to community benchmarks
gpt-oss-20b-MXFP4-Q8
Performance
8k
tokens
1,139
PP tok/s
35.0
TG tok/s
7192
TTFT (ms)
11.9
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 895.0 | 42.3 | 11.7 GB | view |
| 4k | 1,142 | 39.1 | 11.8 GB | view |
| 8k | 1,139 | 35.0 | 11.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 42.3 | 1.00× |
| 2× | 53.9 | 1.27× |
| 4× | 59.6 | 1.41× |
| 8× | 64.5 | 1.52× |