← Back to community benchmarks
Qwen3.6-35B-A3Bmxfp8
Performance
16k
tokens
3,615
PP tok/s
85.9
TG tok/s
4532
TTFT (ms)
36.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
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 | 2,199 | 99.1 | 34.4 GB | view |
| 4k | 3,670 | 94.8 | 35.2 GB | view |
| 8k | 3,876 | 93.7 | 35.5 GB | view |
| 16k | 3,615 | 85.9 | 36.1 GB | current |
| 32k | 3,264 | 76.9 | 37.5 GB | view |
| 64k | 2,272 | 47.5 | 40.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 99.1 | 1.00× |
| 2× | 157.8 | 1.59× |
| 4× | 217.8 | 2.20× |