← Back to community benchmarks
Qwen3.6-35B-A3B-oQ8
Performance
16k
tokens
1,096
PP tok/s
48.0
TG tok/s
14952
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M1 Ultra (64c)
Memory
128 GB
GPU Cores
64
Software
oMLX
v0.3.6
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 825.2 | 55.9 | 35.4 GB | view |
| 4k | 1,098 | 54.1 | 36.2 GB | view |
| 8k | 1,126 | 53.7 | 36.5 GB | view |
| 16k | 1,096 | 48.0 | 37.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 55.9 | 1.00× |
| 2× | 97.6 | 1.75× |
| 4× | 150.3 | 2.69× |