← Back to community benchmarks
Qwen3.5-9B-Gemini-3.1-Pro-Reasoning-Distill
Performance
1k
tokens
208.6
PP tok/s
14.0
TG tok/s
4910
TTFT (ms)
7.8
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.8rc1
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 208.6 | 14.0 | 7.8 GB | current |
| 4k | 207.7 | 12.3 | 8.3 GB | view |
| 8k | 197.7 | 12.3 | 8.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 14.0 | 1.00× |
| 2× | 25.0 | 1.79× |
| 4× | 50.3 | 3.59× |
| 8× | 103.5 | 7.39× |