← Back to community benchmarks
Qwen3.6-28B-REAP
Performance
1k
tokens
915.6
PP tok/s
88.5
TG tok/s
1118
TTFT (ms)
15.9
Peak mem (GB)
Hardware
Chip
M3 Max (30c)
Memory
36 GB
GPU Cores
30
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 915.6 | 88.5 | 15.9 GB | current |
| 4k | 1,122 | 82.8 | 16.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 88.5 | 1.00× |
| 2× | 137.0 | 1.55× |
| 4× | 190.2 | 2.15× |