← Back to community benchmarks
MiniMax-M2.5
Performance
32k
tokens
330.8
PP tok/s
16.1
TG tok/s
99070
TTFT (ms)
181.7
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 496.7 | 40.0 | 174.0 GB | view |
| 4k | 554.8 | 33.9 | 175.0 GB | view |
| 8k | 514.0 | 28.9 | 175.9 GB | view |
| 16k | 436.3 | 23.5 | 177.8 GB | view |
| 32k | 330.8 | 16.1 | 181.7 GB | current |
| 64k | 221.1 | 9.6 | 189.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 40.0 | 1.00× |
| 2× | 56.0 | 1.40× |
| 4× | 71.1 | 1.78× |
| 8× | 87.5 | 2.19× |