← Back to community benchmarks
Qwen3.5-0.8B
Performance
8k
tokens
2,456
PP tok/s
160.8
TG tok/s
3336
TTFT (ms)
2.2
Peak mem (GB)
Hardware
Chip
M1 Max (24c)
Memory
32 GB
GPU Cores
24
Software
oMLX
v0.2.6
macOS
macOS 26.2
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,507 | 199.7 | 2.0 GB | view |
| 4k | 2,544 | 182.8 | 2.2 GB | view |
| 8k | 2,456 | 160.8 | 2.2 GB | current |
| 16k | 2,281 | 123.4 | 2.5 GB | view |
| 32k | 1,878 | 95.8 | 3.1 GB | view |
| 64k | 1,370 | 68.2 | 4.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 199.7 | 1.00× |
| 2× | 275.7 | 1.38× |
| 4× | 364.2 | 1.82× |
| 8× | 403.9 | 2.02× |