← Back to community benchmarks
gemma-3-27b-it
Performance
16k
tokens
379.3
PP tok/s
8.2
TG tok/s
43195
TTFT (ms)
31.9
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.10
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 380.0 | 8.8 | 29.9 GB | view |
| 4k | 413.2 | 9.3 | 30.9 GB | view |
| 8k | 414.9 | 8.9 | 31.3 GB | view |
| 16k | 379.3 | 8.2 | 31.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 8.8 | 1.00× |
| 2× | 12.8 | 1.45× |
| 4× | 19.7 | 2.24× |
| 8× | 22.2 | 2.52× |