← Back to community benchmarks
Qwen3.5-122B-A10B-8bit-abliterated
Performance
1k
tokens
684.7
PP tok/s
17.7
TG tok/s
1496
TTFT (ms)
125.1
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.12
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 684.7 | 17.7 | 125.1 GB | current |
| 4k | 754.0 | 17.4 | 125.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 17.7 | 1.00× |
| 2× | 31.3 | 1.77× |
| 4× | 50.7 | 2.86× |