Parameters
MoE · FP8 · 865 GBOpen Model Catalogue
DeepSeek
Text generationReasoningCoding
DeepSeek V4 Pro
deepseek-ai/DeepSeek-V4-ProDeepSeek newest flagship — 8 × H200
Min VRAM
≈ 950 GBHardware
12 × 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.