Almog's Inference Startups

Open inference infrastructure from the people behind vLLM.

Efficient inference and persistent computers for long-running AI agents.

Frontier inference clusters, co-designed from chips to software.

The AI-native cloud for inference, model training, and compute.

AI infrastructure for training, inference, and agents, scaling to thousands of GPUs.

Memory-centric compute for AI inference.

Train and serve AI models on one stack, with engineers who help you reach production.

AI inference on owned infrastructure, with model APIs, private deployments, and GPU compute.

GPU instances, serverless compute, and model APIs for AI developers.

Own your AI: train and serve open models that become your specialized intelligence.

Open infrastructure for AI inference and training

Open-source model serving and inference infrastructure for deploying AI across clouds with control over performance and cost.

Inference that keeps getting better.

Open compute for AI workloads that refuse to stand still.