← Back to community benchmarks
Qwen3.5-2B
Performance
16k
tokens
3,078
PP tok/s
138.6
TG tok/s
5323
TTFT (ms)
3.2
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
48 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,001 | 173.0 | 2.5 GB | view |
| 4k | 2,886 | 166.3 | 2.8 GB | view |
| 8k | 3,121 | 155.6 | 3.1 GB | view |
| 16k | 3,078 | 138.6 | 3.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 173.0 | 1.00× |
| 2× | 291.5 | 1.68× |
| 4× | 344.4 | 1.99× |
| 8× | 384.1 | 2.22× |