← Back to community benchmarks
MiniMax-M2.79bit
Performance
4k
tokens
744.5
PP tok/s
31.4
TG tok/s
5502
TTFT (ms)
241.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.4
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 613.4 | 35.4 | 240.5 GB | view |
| 4k | 744.5 | 31.4 | 241.5 GB | current |
| 8k | 694.0 | 27.6 | 242.4 GB | view |
| 16k | 591.6 | 22.6 | 244.4 GB | view |
| 32k | 450.3 | 15.3 | 248.2 GB | view |
| 64k | 302.8 | 10.6 | 256.0 GB | view |
| 128k | 180.8 | 6.1 | 271.8 GB | view |
| 195k | 127.1 | 4.4 | 288.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 35.4 | 1.00× |
| 2× | 49.5 | 1.40× |
| 4× | 61.9 | 1.75× |
| 8× | 88.1 | 2.49× |