Companies in the category 'AI Infrastructure'

Prime Intellect provides a full-stack platform for training, deploying, and continuously improving AI models and agents. The platform combines compute orchestration, reinforcement learning tooling, evaluation infrastructure, and open-source research to make frontier AI training accessible to organizations of all sizes. Its offerings include hosted RL training on thousands of open-source environments, one-click inference deployment, and a compute marketplace aggregating GPU resources globally. Prime Intellect also develops and publishes open-source models and frameworks, including prime-rl for large-scale asynchronous RL and verifiers for building RL environments.

Bespoke Labs is an applied AI research lab that builds reinforcement learning environments and data curation infrastructure for training and evaluating long-horizon AI agents. The company develops open-source tools including Curator, a synthetic data curation framework, and contributes to benchmarks such as Terminal-Bench and OpenThoughts to advance the reliability of AI agents in production environments.

Modular is an AI infrastructure company that builds a unified compute platform for developing and deploying generative AI applications. Its products include MAX, a high-performance AI inference engine, and Mojo, a Python-compatible programming language designed for high-performance AI and systems programming on CPUs and GPUs.

Blaxel is a cloud infrastructure platform purpose-built for AI agents, providing persistent sandbox environments that automatically scale to zero after one second of inactivity and resume in under 25ms with full memory state preserved. The platform co-locates agent logic alongside sandboxes to eliminate network overhead and deliver near-instant latency. Blaxel also offers serverless agent hosting and an intelligent LLM gateway, enabling developers to maintain millions of secure sandboxes on standby indefinitely.

dstack is an open-source control plane for provisioning compute and running AI workloads across GPU clouds, Kubernetes, and bare-metal clusters. It supports NVIDIA, AMD, TPU, and other accelerators, and provides primitives for dev environments, training tasks, inference services, and fleets. dstack is designed for both engineers and AI agents, offering a unified interface that eliminates the need for Kubernetes or Slurm expertise.

RadixArk is an infrastructure-first company building large-scale AI inference and training systems. It is the commercial entity behind SGLang, an open-source inference engine originally developed at UC Berkeley, and Miles, an enterprise-facing reinforcement learning framework for LLM and VLM post-training. The company offers hosted inference services built on top of its open-source tooling, targeting AI teams that need fast, cost-efficient model serving at scale.

Inferact is a startup founded by the creators and core maintainers of vLLM, the most widely adopted open-source LLM inference engine. The company's mission is to grow vLLM as the world's AI inference engine and accelerate AI progress by making inference cheaper and faster.

Daytona provides secure, elastic infrastructure for running AI-generated code, enabling developers and AI agents to execute code in isolated sandbox environments. Founded in 2023 by the team behind Codeanywhere, the platform delivers sub-90ms sandbox creation, stateful environments with persistent state, and massive parallelization for concurrent AI agent workflows. Daytona supports programmatic control via File, Git, LSP, and Execute APIs, and is designed to meet enterprise compliance requirements including HIPAA, SOC 2, and GDPR.