← Back to community benchmarks
Qwen3-Coder-Next
Performance
16k
tokens
1,175
PP tok/s
60.7
TG tok/s
13945
TTFT (ms)
43.9
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.7
macOS
macOS 26.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,109 | 83.5 | 43.0 GB | view |
| 4k | 1,287 | 77.7 | 43.2 GB | view |
| 8k | 1,269 | 73.7 | 43.4 GB | view |
| 16k | 1,175 | 60.7 | 43.9 GB | current |
| 32k | 976.8 | 42.6 | 44.7 GB | view |
| 64k | 759.3 | 22.7 | 46.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 83.5 | 1.00× |
| 2× | 137.3 | 1.64× |
| 4× | 185.7 | 2.22× |
| 8× | 266.8 | 3.20× |