← Back to community benchmarks
6cee5e81ee83917806bbde320786a8fb61efebee symlink
Performance
4k
tokens
1,841
PP tok/s
96.5
TG tok/s
2225
TTFT (ms)
13.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
96 GB
GPU Cores
60
Software
oMLX
v0.3.0
macOS
macOS 26.2
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,690 | 103.8 | 13.3 GB | view |
| 4k | 1,841 | 96.5 | 13.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 103.8 | 1.00× |
| 2× | 102.0 | 0.98× |
| 4× | 164.5 | 1.58× |