Crusoe Managed Inference

What it does

Crusoe Managed Inference serves open-source large language and generative models through managed endpoints. Users send prompts to a model and receive generated outputs. The service is accessed through Crusoe Intelligence Foundry, where API keys can be generated and performance metrics monitored.

Three deployment options are offered. Serverless Inference is a fully managed inference API with usage-based consumption. Self-Serve Deployments run top models on dedicated endpoints tuned for throughput or responsiveness and can serve models customized with Serverless Fine-Tuning. Tailored Deployments are dedicated, benchmarked endpoints built by the Crusoe team around a customer-supplied model.

It suits developers and teams building AI applications who want managed open-model serving, from early experimentation to production-scale workloads. Access starts with a request through the Crusoe console. Serverless pricing varies by model and token type, while Self-Serve monthly and volume rates require contacting sales.

What it can do

• Serverless inference over a curated selection of open-source models with pay-as-you-go token pricing
• Self-Serve Deployments offering dedicated endpoints optimized for throughput or responsiveness
• Tailored Deployments with a dedicated, benchmarked endpoint for customer-supplied models
• Serverless Fine-Tuning and deployment of fine-tuned models in one click
• API key generation and performance monitoring in Crusoe Intelligence Foundry
• Provisioned throughput via AI Model Units for guaranteed throughput commitments

Works with

• Crusoe Cloud Console
• Crusoe Intelligence Foundry

Before you choose

• Serverless Inference is billed per token; monthly and volume rates for Self-Serve Deployments and Provisioned Throughput require contacting sales.
• Crusoe's published speedup figures (up to 9.9x time-to-first-token and 5x throughput) are benchmarked against vLLM on Llama-3.3-70B in a 4-node deployment, so results on other models may differ.
• Model availability varies by deployment option; the model hub lists which models support Serverless Inference versus Self-Serve Deployments.

Listed under

• Coding
• Sandboxes and infrastructure
• Developers
• Fine-tuning
• GPU deployment
• Inference
• LLM API
• Open models

Get started

Start by requesting access to Crusoe Intelligence Foundry; a free trial is offered on the product page.

1. Request access to Crusoe Intelligence Foundry through the console request page.
2. Generate an API key in Crusoe Intelligence Foundry for programmatic use.
3. Choose a deployment option (Serverless Inference, Self-Serve Deployments, or Tailored Deployments) based on workload and read the linked documentation.

Serverless Inference is billed per 1M tokens, with separate input, output, and cached token rates that vary by model. Self-Serve Deployments are billed per GPU-hour, with monthly and volume rates available through sales. Confirm current model-specific rates on the pricing page before budgeting.

Source: crusoe.ai/cloud/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Managed inference for open models, low latency | Crusoe ↗www.crusoe.ai
• GPU pricing for AI compute & inference | Crusoe ↗www.crusoe.ai
• Crusoe | The AI factory company ↗www.crusoe.ai

Company website

crusoe.ai: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.