Parameters
230B MoE · 10B active · FP8Open Model Catalogue
MiniMax
CodingText generationReasoning
MiniMax M2.5
MiniMaxAI/MiniMax-M2.5Agentic coding MoE — 2 × H200
Min VRAM
≈ 260 GBHardware
4 × 80 GB GPUOpenAI-compatible API with a per-workload key — drop-in for existing SDKs.
Dedicated GPUs — your model, your instance, no shared queues.
Billed by the second with auto-shutdown — never pay for idle time you didn't plan.