← Back to community benchmarks
Qwen3___5-27B-Claude-4___6-OS-INSTRUCT-mxfp8
Performance
4k
tokens
244.9
PP tok/s
16.2
TG tok/s
16724
TTFT (ms)
28.7
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 239.7 | 16.6 | 28.4 GB | view |
| 4k | 244.9 | 16.2 | 28.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.6 | 1.00× |
| 2× | 26.2 | 1.58× |
| 4× | 27.9 | 1.68× |