← Back to community benchmarks
Qwen3-Coder-Next
Performance
4k
tokens
1,820
PP tok/s
60.0
TG tok/s
2251
TTFT (ms)
80.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.10
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,296 | 63.1 | 80.1 GB | view |
| 4k | 1,820 | 60.0 | 80.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 63.1 | 1.00× |
| 2× | 107.7 | 1.71× |
| 4× | 172.3 | 2.73× |