← Back to community benchmarks
Qwen3.5-0.8B
Performance
4k
tokens
4,363
PP tok/s
199.3
TG tok/s
939
TTFT (ms)
2.2
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.6
macOS
macOS 26.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 4,307 | 208.9 | 2.0 GB | view |
| 4k | 4,363 | 199.3 | 2.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 208.9 | 1.00× |
| 2× | 221.2 | 1.06× |
| 4× | 470.1 | 2.25× |
| 8× | 529.6 | 2.54× |