← Back to community benchmarks
Qwen3.5-9B-Claude-4.6-HighIQ-INSTRUCT-HERETIC-UNCENSOREDmxfp8
Performance
4k
tokens
144.3
PP tok/s
19.0
TG tok/s
28390
TTFT (ms)
11.3
Peak mem (GB)
Hardware
Chip
M1 Pro (14c)
Memory
32 GB
GPU Cores
14
Software
oMLX
v0.2.20
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 143.8 | 19.7 | 10.5 GB | view |
| 4k | 144.3 | 19.0 | 11.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.7 | 1.00× |
| 2× | 22.3 | 1.13× |
| 4× | 22.2 | 1.13× |