QriticalBeta
Open Model Catalogue
C
Cohere
Text generationReasoning

Command A+

CohereLabs/command-a-plus-05-2026-fp8

Cohere flagship · Apache 2.0 — multi-GPU

Parameters
218B MoE
Min VRAM
≈ 260 GB
Hardware
4 × 80 GB GPU
This model requires accepting its licence on HuggingFace and a HuggingFace token in your account credentials.
OpenAI-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.