← Back to community benchmarks
Qwen3.5-4B
Performance
16k
tokens
254.5
PP tok/s
24.5
TG tok/s
64371
TTFT (ms)
4.8
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
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 | 283.0 | 34.0 | 3.9 GB | view |
| 4k | 279.6 | 32.3 | 4.2 GB | view |
| 8k | 267.9 | 29.4 | 4.6 GB | view |
| 16k | 254.5 | 24.5 | 4.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 34.0 | 1.00× |
| 2× | 51.1 | 1.50× |
| 4× | 58.8 | 1.73× |