Vast.ai Serverless deploys models as autoscaling inference endpoints, with optimization that selects hardware across the Vast.ai fleet and scales to zero between requests. Deployments are defined in Python through the Vast SDK.
Buyers pay only for compute time while handling inference traffic, without managing GPU capacity, tiers, or minimum commitments.

Vast.ai Serverless runs autoscaling GPU inference endpoints on managed capacity.


Baseten Dedicated Inference is a high-performance inference platform for deploying, serving, and scaling custom and open-source AI models in production.


Modal is a high-performance AI infrastructure cloud for online, batch, and multi-modal inference with instant autoscaling.
Baseten Dedicated Inference is a high-performance inference platform for deploying, serving, and scaling custom and open-source AI models in production.

Modal is a high-performance AI infrastructure cloud for online, batch, and multi-modal inference with instant autoscaling.



Vultr Serverless Inference deploys and serves generative AI models for businesses through managed GPU infrastructure.



Together AI is a serverless and dedicated inference API for open-source AI models with OpenAI-compatible endpoints.


RunInfra Model APIs provide hosted open-model inference through compatible APIs for developers and agent teams.


Async inference API for long-horizon AI agents, serving leading open-source models through OpenAI- and Anthropic-compatible endpoints.


Vultr Cloud GPU supplies on-demand and dedicated GPU infrastructure to businesses for accelerated computing.


GMI Cloud Inference is a Model-as-a-Service platform serving hosted LLM, image, video, and audio models through one unified API.


Fastest cloud for training/inference of 1B-24T models with no sharding.


Lambda Cloud is Lambda's public GPU cloud offering on-demand instances, reserved clusters, and single-tenant supercomputing capacity.


One HIPAA-compliant gateway routing all LLM traffic with BAA coverage, audit logging, key management, and cost controls enforced automatically.



Managed cloud execution service running durable AI agents and background tasks with retries, checkpointing, monitoring, and elastic scaling.


Managed service for fine-tuning open-source models on dedicated or shared GPU infrastructure for production use.


Persistent hosted Linux sandboxes for long-horizon agents, billed per actively-used compute, with pause, resume, and fork.


Cloudflare's managed gateway for routing, securing, caching, and observing AI applications across many model providers.


Fireworks AI is a managed inference and fine-tuning platform for open-source generative AI models.


Hosted inference API for the ESM model families with credit-based metering and a developer console.


Agent 37 Cloud API gives developers a programmable way to run persistent, isolated AI agent sandboxes.

Vultr Serverless Inference deploys and serves generative AI models for businesses through managed GPU infrastructure.

Replicate is a cloud API to run, fine-tune, and deploy open-source machine learning models with auto-scaling.

Together AI is a serverless and dedicated inference API for open-source AI models with OpenAI-compatible endpoints.

RunInfra Model APIs provide hosted open-model inference through compatible APIs for developers and agent teams.

Async inference API for long-horizon AI agents, serving leading open-source models through OpenAI- and Anthropic-compatible endpoints.

Vultr Cloud GPU supplies on-demand and dedicated GPU infrastructure to businesses for accelerated computing.

GMI Cloud Inference is a Model-as-a-Service platform serving hosted LLM, image, video, and audio models through one unified API.

Fastest cloud for training/inference of 1B-24T models with no sharding.

Lambda Cloud is Lambda's public GPU cloud offering on-demand instances, reserved clusters, and single-tenant supercomputing capacity.

One HIPAA-compliant gateway routing all LLM traffic with BAA coverage, audit logging, key management, and cost controls enforced automatically.

Containerized inference microservices providing prebuilt APIs for popular AI models on NVIDIA accelerated infrastructure.

Managed cloud execution service running durable AI agents and background tasks with retries, checkpointing, monitoring, and elastic scaling.

Managed service for fine-tuning open-source models on dedicated or shared GPU infrastructure for production use.

Persistent hosted Linux sandboxes for long-horizon agents, billed per actively-used compute, with pause, resume, and fork.

Cloudflare's managed gateway for routing, securing, caching, and observing AI applications across many model providers.

Fireworks AI is a managed inference and fine-tuning platform for open-source generative AI models.

Hosted inference API for the ESM model families with credit-based metering and a developer console.

Agent 37 Cloud API gives developers a programmable way to run persistent, isolated AI agent sandboxes.

A unified AI gateway and LLM router exposing one OpenAI-compatible endpoint across 600+ models.
