← Back to community benchmarks
Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16-source-oQ6-mtp
Performance
1k
tokens
224.4
PP tok/s
21.9
TG tok/s
4563
TTFT (ms)
23.0
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 224.4 | 21.9 | 23.0 GB | current |
| 4k | 181.5 | 20.2 | 24.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.9 | 1.00× |
| 2× | 22.0 | 1.00× |
| 4× | 24.3 | 1.11× |