← Back to community benchmarks
Qwen3-Coder-Next
Performance
8k
tokens
2,078
PP tok/s
62.8
TG tok/s
3943
TTFT (ms)
81.4
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
256 GB
GPU Cores
80
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,576 | 66.8 | 80.1 GB | view |
| 4k | 2,058 | 65.3 | 81.0 GB | view |
| 8k | 2,078 | 62.8 | 81.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 66.8 | 1.00× |
| 2× | 116.3 | 1.74× |
| 4× | 179.1 | 2.68× |
| 8× | 264.8 | 3.96× |