← Back to community benchmarks
Qwen3.5-27B-Claude-4.6-OS-INSTRUCT-mxfp8
Performance
4k
tokens
187.3
PP tok/s
19.0
TG tok/s
21873
TTFT (ms)
30.0
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.2.18
macOS
macOS 15.7.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 185.2 | 20.1 | 28.4 GB | view |
| 4k | 187.3 | 19.0 | 30.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.1 | 1.00× |
| 2× | 26.3 | 1.31× |
| 4× | 26.9 | 1.34× |