← Back to community benchmarks
Carnice-V2-27b-oQ6
Performance
4k
tokens
244.6
PP tok/s
19.9
TG tok/s
16748
TTFT (ms)
24.0
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 236.2 | 20.3 | 22.6 GB | view |
| 4k | 244.6 | 19.9 | 24.0 GB | current |
| 16k | 236.7 | 18.9 | 26.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.3 | 1.00× |
| 2× | 33.2 | 1.64× |
| 4× | 36.9 | 1.82× |