← Back to community benchmarks
Qwen3-Coder-Next-oQ8e
Performance
32k
tokens
534.9
PP tok/s
36.9
TG tok/s
61262
TTFT (ms)
81.9
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.1
macOS
macOS 26.5.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 432.5 | 37.6 | 80.2 GB | view |
| 4k | 583.7 | 37.9 | 80.9 GB | view |
| 8k | 562.5 | 36.1 | 81.2 GB | view |
| 16k | 599.2 | 52.2 | 81.4 GB | view |
| 32k | 534.9 | 36.9 | 81.9 GB | current |
| 64k | 446.7 | 32.2 | 82.7 GB | view |
| 128k | 252.2 | 23.6 | 84.0 GB | view |
| 195k | 188.9 | 20.6 | 85.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 37.6 | 1.00× |
| 2× | 69.1 | 1.84× |
| 4× | 100.4 | 2.67× |
| 8× | 126.8 | 3.37× |