Provider directory

Compare AI infrastructure providers

Browse local provider profiles covering category, best-fit workloads, pricing style, regions, complexity and enterprise readiness.

OpenRouter

Inference API routing

Unified API access to multiple model providers through a routing layer.

Best for: Model experimentation, Inference API routing, Fast prototype switching
Pricing style: Token-based pass-through and routing fees
Complexity: Low
Enterprise: Moderate
Regions: Global API access, Provider-dependent residency

View provider

RunPod

GPU cloud

GPU instances and serverless GPU options for development and inference workloads.

Best for: GPU development, Cost-sensitive experiments, Custom model hosting
Pricing style: Hourly GPU and serverless pricing
Complexity: Medium
Enterprise: Moderate
Regions: Global capacity varies by GPU type

View provider

Vast AI

GPU marketplace

Marketplace-style access to distributed GPU machines from independent hosts.

Best for: Low-cost GPU experiments, Batch jobs, Flexible GPU access
Pricing style: Marketplace hourly GPU pricing
Complexity: High
Enterprise: Emerging
Regions: Host-dependent global availability

View provider

Together AI

Managed inference and training

Managed inference, fine-tuning and dedicated infrastructure for open models.

Best for: Open model inference, Fine-tuning, Enterprise open-model deployments
Pricing style: Token-based, fine-tuning and dedicated deployment pricing
Complexity: Low
Enterprise: High
Regions: US-focused and enterprise-dependent options

View provider

Fireworks AI

Inference API

Fast managed inference for open and fine-tuned models.

Best for: Low-latency inference, Open model APIs, Production model serving
Pricing style: Token-based and dedicated inference pricing
Complexity: Low
Enterprise: High
Regions: Provider-managed regions

View provider

Lambda Labs

GPU cloud

GPU cloud instances and infrastructure for machine learning teams.

Best for: Dedicated GPU instances, Training workloads, ML infrastructure teams
Pricing style: Hourly GPU instance pricing and reserved capacity
Complexity: Medium
Enterprise: Moderate
Regions: US and selected global capacity

View provider

AWS GPU Instances

Hyperscale cloud GPU

GPU instances and managed AI services inside the AWS cloud ecosystem.

Best for: Enterprise infrastructure, Compliance-heavy deployments, Existing AWS teams
Pricing style: On-demand, reserved and savings-plan infrastructure pricing
Complexity: High
Enterprise: High
Regions: Broad global AWS regions with capacity differences

View provider

Google Cloud GPU

Hyperscale cloud GPU

GPU infrastructure and managed AI services in Google Cloud.

Best for: Google Cloud teams, Enterprise AI platforms, Large-scale ML workflows
Pricing style: Cloud infrastructure pricing and managed service pricing
Complexity: High
Enterprise: High
Regions: Broad Google Cloud regions with capacity differences

View provider

Azure AI / GPU

Hyperscale cloud GPU

Microsoft cloud GPU infrastructure and Azure AI services.

Best for: Microsoft enterprise environments, Governed AI, Private deployments
Pricing style: Cloud infrastructure, managed AI and committed capacity pricing
Complexity: High
Enterprise: High
Regions: Broad Azure regions with capacity differences

View provider

Modal

Serverless compute

Developer-focused serverless compute for AI jobs, inference and data workflows.

Best for: Python-native AI apps, Serverless GPU jobs, Rapid internal tooling
Pricing style: Usage-based serverless compute pricing
Complexity: Medium
Enterprise: Moderate
Regions: Provider-managed regions

View provider

Replicate

Model API platform

API access for running and publishing open models with managed infrastructure.

Best for: Model demos, API-based inference, Open model experimentation
Pricing style: Usage-based model runtime pricing
Complexity: Low
Enterprise: Moderate
Regions: Provider-managed regions

View provider

Groq

Inference API

High-speed inference API built around specialized inference hardware.

Best for: Latency-sensitive inference, Hosted model APIs, Developer prototypes
Pricing style: Token-based inference pricing
Complexity: Low
Enterprise: Moderate
Regions: Provider-managed regions

View provider

OpenAI

Foundation model API

Managed API access to proprietary frontier and production model capabilities.

Best for: Managed LLM APIs, Product teams, High-level AI capabilities
Pricing style: Token-based API pricing and enterprise contracts
Complexity: Low
Enterprise: High
Regions: Provider-managed regions and enterprise-dependent options

View provider

Anthropic

Foundation model API

Managed API access to Claude models for enterprise and developer workloads.

Best for: Managed LLM APIs, Enterprise assistant use cases, Text-heavy workloads
Pricing style: Token-based API pricing and enterprise contracts
Complexity: Low
Enterprise: High
Regions: Provider-managed regions and partner cloud options

View provider

Mistral AI

Model provider

European AI model provider with managed APIs and open model options.

Best for: European AI strategy, Open model options, Managed inference
Pricing style: Token-based API and enterprise deployment pricing
Complexity: Low
Enterprise: High
Regions: EU-oriented and provider-managed options

View provider