← Back to community benchmarks
Qwen3.5-4B
Performance
16k
tokens
414.2
PP tok/s
43.2
TG tok/s
39560
TTFT (ms)
5.3
Peak mem (GB)
Hardware
Chip
M1 Max (24c)
Memory
64 GB
GPU Cores
24
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 435.7 | 73.3 | 4.0 GB | view |
| 4k | 440.0 | 66.9 | 4.2 GB | view |
| 8k | 430.6 | 57.4 | 4.6 GB | view |
| 16k | 414.2 | 43.2 | 5.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 73.3 | 1.00× |
| 2× | 84.7 | 1.16× |
| 4× | 96.0 | 1.31× |
| 8× | 76.3 | 1.04× |