← Back to community benchmarks
Qwen3-Coder-Next
Performance
16k
tokens
956.5
PP tok/s
47.0
TG tok/s
17129
TTFT (ms)
62.4
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
128 GB
GPU Cores
76
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 751.3 | 57.9 | 61.6 GB | view |
| 4k | 966.4 | 56.2 | 61.7 GB | view |
| 8k | 986.0 | 53.4 | 62.0 GB | view |
| 16k | 956.5 | 47.0 | 62.4 GB | current |
| 32k | 881.3 | 39.1 | 63.3 GB | view |
| 64k | 731.4 | 28.9 | 65.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 57.9 | 1.00× |
| 2× | 98.1 | 1.69× |
| 4× | 145.6 | 2.51× |
| 8× | 199.4 | 3.44× |