← Back to community benchmarks
Laguna-XS-2.1
Performance
1k
tokens
1,740
PP tok/s
87.6
TG tok/s
589
TTFT (ms)
33.7
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.5.3
macOS
macOS 26.5.2
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,740 | 87.6 | 33.7 GB | current |
| 4k | 1,422 | 84.0 | 33.8 GB | view |
| 8k | 1,306 | 81.2 | 34.0 GB | view |
| 16k | 1,179 | 76.9 | 34.3 GB | view |
| 32k | 1,012 | 70.5 | 35.0 GB | view |
| 64k | 790.3 | 59.7 | 36.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 87.6 | 1.00× |
| 2× | 140.7 | 1.61× |
| 4× | 223.9 | 2.56× |
| 8× | 314.3 | 3.59× |