← Back to community benchmarks
DeepSeek-R1-0528-Qwen3-8B
Performance
4k
tokens
307.3
PP tok/s
46.6
TG tok/s
13330
TTFT (ms)
5.7
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.8
macOS
macOS 26.0.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 304.4 | 56.1 | 5.0 GB | view |
| 4k | 307.3 | 46.6 | 5.7 GB | current |
| 8k | 291.2 | 37.7 | 6.6 GB | view |
| 16k | 260.6 | 27.4 | 8.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 56.1 | 1.00× |
| 2× | 66.0 | 1.18× |
| 4× | 72.7 | 1.30× |
| 8× | 64.7 | 1.15× |