← Back to community benchmarks
MiniMax-M2___5
Performance
64k
tokens
298.3
PP tok/s
11.4
TG tok/s
219695
TTFT (ms)
136.4
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 628.4 | 46.0 | 120.7 GB | view |
| 4k | 742.2 | 39.4 | 121.7 GB | view |
| 8k | 701.0 | 33.3 | 122.6 GB | view |
| 16k | 591.6 | 26.3 | 124.6 GB | view |
| 32k | 446.8 | 18.1 | 128.5 GB | view |
| 64k | 298.3 | 11.4 | 136.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.0 | 1.00× |
| 2× | 70.2 | 1.53× |
| 4× | 102.5 | 2.23× |