← Back to community benchmarks
Qwen3-Coder-Next-oQ8e
Performance
8k
tokens
1,160
PP tok/s
63.0
TG tok/s
7063
TTFT (ms)
81.2
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.1
macOS
macOS 26.5.2
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,230 | 65.8 | 80.2 GB | view |
| 4k | 1,288 | 64.2 | 80.9 GB | view |
| 8k | 1,160 | 63.0 | 81.2 GB | current |
| 16k | 998.5 | 59.8 | 81.6 GB | view |
| 32k | 848.5 | 55.8 | 82.2 GB | view |
| 64k | 531.0 | 22.2 | 82.6 GB | view |
| 128k | 257.9 | 23.7 | 84.0 GB | view |
| 195k | 201.5 | 19.9 | 85.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 65.8 | 1.00× |
| 2× | 70.4 | 1.07× |
| 4× | 91.3 | 1.39× |
| 8× | 123.0 | 1.87× |