Thunder Compute

RunPod is a cloud platform for AI and machine learning workloads that rents GPU compute by the second. It offers Pods for persistent GPU instances, Serverless endpoints that autoscale inference from zero, and multi-node Clusters, plus a Hub of ready-to-deploy open-source models and templates.

Miễn phí / Mã nguồn mởĐám mây AI do đội ngũ tạo ra PyTorch Lightning xây dựng

Miễn phí / Mã nguồn mởHạ tầng đám mây serverless cho khối lượng công việc AI và dữ liệu

Thuê GPU đám mây theo từng giây từ chợ toàn cầu

Nebius is an AI-focused cloud platform that gives teams non-virtualized NVIDIA GPU clusters, managed MLOps tooling, and a serverless inference layer so they can move from experiment to production on one stack.

FluidStack builds and operates large-scale compute infrastructure for AI, including gigawatt-class data centers and GPU clusters used for training and inference. The company designs its own power, cooling and data center systems, onshores manufacturing and deploys solar generation to supply the power its clusters need. Rather than a self-serve cloud console, FluidStack works with AI labs and enterprises to provision large reserved clusters, with orchestration tooling for the ML frameworks and schedulers those teams already use.

Massed Compute is a cloud GPU provider offering on-demand NVIDIA instances, single-tenant bare metal servers, large InfiniBand node clusters, and a programmatic REST API and MCP server. It targets AI training, inference, machine learning, data analytics, and VFX rendering workloads.

Miễn phí / Mã nguồn mởGPU serverless cho AI thời gian thực với cold start vài giây

Đám mây AI riêng cho huấn luyện và suy luận

Miễn phí / Mã nguồn mởMột cloud cho hơn 200 API mô hình, agent sandbox và sức mạnh GPU

Miễn phí / Mã nguồn mởĐám mây GPU phân tán điều phối các GPU tiêu dùng nhàn rỗi trên toàn cầu

Miễn phí / Mã nguồn mởGeneral Compute is an inference cloud built around alternative AI silicon rather than standard GPU fleets. The company buys racks of accelerators, racks them in colocation data centres, and handles the cooling, power, and networking itself. On top of that hardware it runs a disaggregated inference stack that splits each request: prompt processing runs on NVIDIA GPUs for prefill, while token generation runs on faster decode chips such as Cerebras, SambaNova, Positron AI, and d-Matrix. Customers get bare metal access or a tokens-per-second SLA.