← Back to community benchmarks
Qwen3.5-9B-Claude-4.6-HighIQ-INSTRUCT-HERETIC-UNCENSOREDmxfp8
Performance
4k
tokens
164.0
PP tok/s
19.5
TG tok/s
24976
TTFT (ms)
10.3
Peak mem (GB)
Hardware
Chip
M1 Pro (16c)
Memory
16 GB
GPU Cores
16
Software
oMLX
v0.3.8rc1
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 157.5 | 20.1 | 9.7 GB | view |
| 4k | 164.0 | 19.5 | 10.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.1 | 1.00× |
| 2× | 25.4 | 1.26× |
| 4× | 26.4 | 1.31× |