← Back to community benchmarks
Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
Performance
128k
tokens
643.0
PP tok/s
22.8
TG tok/s
203852
TTFT (ms)
46.5
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.5.dev1
macOS
macOS 26.4.1
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,316 | 87.3 | 35.4 GB | view |
| 4k | 1,676 | 84.9 | 36.2 GB | view |
| 8k | 1,657 | 78.3 | 36.5 GB | view |
| 16k | 1,570 | 71.9 | 37.1 GB | view |
| 32k | 1,268 | 49.8 | 38.5 GB | view |
| 64k | 992.1 | 33.5 | 41.2 GB | view |
| 128k | 643.0 | 22.8 | 46.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 87.3 | 1.00× |
| 2× | 130.4 | 1.49× |
| 4× | 203.0 | 2.33× |