← Back to community benchmarks
gpt-oss-20b
Performance
8k
tokens
784.2
PP tok/s
47.1
TG tok/s
10446
TTFT (ms)
13.4
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.5
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 711.1 | 53.3 | 13.3 GB | view |
| 4k | 798.4 | 50.5 | 13.3 GB | view |
| 8k | 784.2 | 47.1 | 13.4 GB | current |
| 16k | 720.9 | 41.9 | 13.6 GB | view |
| 32k | 600.1 | 34.3 | 14.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 53.3 | 1.00× |
| 2× | 54.2 | 1.02× |
| 4× | 79.7 | 1.50× |