MiniMax-M2
by MiniMaxAI
229B params · text-generation · 1.5k likes · 211.5k downloads
MiniMax-M2 is a 229B parameter model. At Q4 quantization it requires 114GB of VRAM. It runs comfortably on H200 SXM 141 GB (30 tok/s).
Inference providers
| Provider | $/1M in | $/1M out | Throughput |
|---|---|---|---|
| Novita | 31 tok/s |
GPU compatibility
| GPU | VRAM | Q4 Decode | Verdict |
|---|---|---|---|
| H200 SXM 141 GB | 141GB | 30 tok/s | comfortable |
| M2 Ultra 192GB | 192GB | 2 tok/s | tight |