← Back to community benchmarks
Qwen3.5-4B-oQ4-mtp
Performance
8k
tokens
4,408
PP tok/s
143.9
TG tok/s
1859
TTFT (ms)
4.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.1
macOS
macOS 26.5.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,532 | 146.8 | 3.6 GB | view |
| 4k | 4,112 | 146.5 | 4.3 GB | view |
| 8k | 4,408 | 143.9 | 4.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 146.8 | 1.00× |
| 2× | 222.0 | 1.51× |
| 4× | 293.3 | 2.00× |