← Back to community benchmarks
Qwen3.5-4B
Performance
16k
tokens
367.6
PP tok/s
29.8
TG tok/s
44567
TTFT (ms)
7.7
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 26.2
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 405.7 | 37.4 | 6.7 GB | view |
| 4k | 403.4 | 36.7 | 7.0 GB | view |
| 8k | 391.3 | 34.5 | 7.4 GB | view |
| 16k | 367.6 | 29.8 | 7.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 37.4 | 1.00× |
| 2× | 74.5 | 1.99× |
| 4× | 93.1 | 2.49× |
| 8× | 94.9 | 2.54× |