← Back to community benchmarks
Qwen3-Coder-Next-qx64n-hi
Performance
4k
tokens
1,622
PP tok/s
66.5
TG tok/s
2526
TTFT (ms)
56.6
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
96 GB
GPU Cores
60
Software
oMLX
v0.2.19
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,313 | 69.1 | 55.6 GB | view |
| 4k | 1,622 | 66.5 | 56.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 69.1 | 1.00× |
| 2× | 118.0 | 1.71× |
| 4× | 180.1 | 2.61× |
| 8× | 262.8 | 3.80× |