← Back to community benchmarks
Qwen3.5-4B
Performance
1k
tokens
560.9
PP tok/s
86.3
TG tok/s
1826
TTFT (ms)
4.0
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 560.9 | 86.3 | 4.0 GB | current |
| 4k | 571.4 | 80.6 | 4.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 86.3 | 1.00× |
| 2× | 105.3 | 1.22× |
| 4× | 118.3 | 1.37× |