← Back to community benchmarks
MiniMax-M2.7-RAM-120GB
Performance
32k
tokens
456.1
PP tok/s
17.3
TG tok/s
71840
TTFT (ms)
120.4
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 650.9 | 44.3 | 112.7 GB | view |
| 8k | 715.0 | 31.9 | 114.7 GB | view |
| 16k | 604.9 | 24.7 | 116.5 GB | view |
| 32k | 456.1 | 17.3 | 120.4 GB | current |
| 64k | 300.3 | 10.8 | 128.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 44.3 | 1.00× |
| 4× | 85.1 | 1.92× |