← Back to community benchmarks
Kimi-K2.63.6bit
Performance
1k
tokens
208.0
PP tok/s
20.3
TG tok/s
4922
TTFT (ms)
428.8
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 208.0 | 20.3 | 428.8 GB | current |
| 4k | 238.1 | 20.4 | 430.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.3 | 1.00× |
| 2× | 29.4 | 1.45× |
| 4× | 37.6 | 1.85× |