← Back to community benchmarks
Jundot-Qwen3.6-27B-oQ4-mtp
Performance
4k
tokens
92.4
PP tok/s
18.5
TG tok/s
44328
TTFT (ms)
17.9
Peak mem (GB)
Hardware
Chip
M2 Max (30c)
Memory
64 GB
GPU Cores
30
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 90.6 | 19.3 | 16.5 GB | view |
| 4k | 92.4 | 18.5 | 17.9 GB | current |
| 8k | 91.9 | 17.8 | 18.9 GB | view |
| 16k | 90.2 | 16.5 | 20.4 GB | view |
| 32k | 86.8 | 15.1 | 23.4 GB | view |
| 64k | 80.1 | 12.6 | 29.5 GB | view |
| 128k | 69.3 | 9.4 | 41.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.3 | 1.00× |
| 2× | 21.5 | 1.11× |
| 4× | 23.2 | 1.20× |