← Back to community benchmarks
Qwen3.5-0.8B
Performance
1k
tokens
1,625
PP tok/s
17.6
TG tok/s
630
TTFT (ms)
3.5
Peak mem (GB)
Hardware
Chip
M3 Pro (18c)
Memory
18 GB
GPU Cores
18
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,625 | 17.6 | 3.5 GB | current |
| 4k | 2,235 | 17.6 | 3.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 17.6 | 1.00× |
| 2× | 30.1 | 1.71× |
| 4× | 34.0 | 1.93× |
| 8× | 52.5 | 2.98× |