- Home
- LLM gateways
- LiteLLM
LiteLLM
What it is
LiteLLM is an open-source AI gateway and LLM proxy for platform teams. It gives applications a shared OpenAI-compatible API to 140+ providers and 1,800+ models. Routing options include load balancing, selecting lower-cost routes, and sending simple prompts automatically to cheaper models. Teams can track usage and spend by key, user, team, organization, tool, agent, and MCP, with hard budgets and rate limits. Response and semantic caching can use Redis, S3, or GCS. LiteLLM can be self-hosted, including in air-gapped environments, and offers deployment through an official Helm chart or Terraform module. Its self-hosted deployment is described as having no telemetry and keeping data within the user's infrastructure. The free Open Source plan is free, billed Free forever, and includes self-hosting, 100+ providers, virtual keys, spend tracking, budgets, rate limits, fallbacks, logging, and Prometheus metrics. A 30-day Enterprise trial key is available without a credit card.
For and against
For
- One API connects to 140+ providers and 1,800+ models.
- Offers load balancing and lower-cost routing.
- Tracks spend and supports budgets and rate limits.
- Self-hosting can keep data within your infrastructure.
Against
- Enterprise plan is billed annually.
- Enterprise trial lasts 30 days.
The verdict
LiteLLM combines provider access, routing, spend controls, and self-hosting options in one gateway. The free Open Source plan is self-hosted; Enterprise pricing and deployment-specific costs require a separate inquiry.
Who it suits
LiteLLM suits platform teams that need a shared API for multiple model providers and controls for usage, routing, and spend. Its Enterprise plan is described for production teams, including high-volume and regulated deployments.
Plans and prices
Open Source
100+ providers · virtual keys, users and teams · spend tracking · budgets and rate limits
Read 2 Oct 2026 · litellm.ai
Enterprise
Annual request capacity · priced by deployment architecture and support needs · volume discounts · 24/7 support and SLAs
Read 2 Oct 2026 · litellm.ai
On the record
- Caching
- Response and semantic caching can use Redis, S3, or GCS.litellm.ai · read 2 Oct 2026
- Data handling
- LiteLLM says its self-hosted deployment has no telemetry and keeps data within the user's infrastructure.litellm.ai · read 2 Oct 2026
- Deployment
- The gateway can be self-hosted, including in air-gapped environments, and supports deployment with an official Helm chart or Terraform module.litellm.ai · read 2 Oct 2026
- Enterprise access
- The Enterprise plan includes SSO and SCIM, OIDC/JWT authentication, audit logs, secret managers, key rotation, and organization and team administrators.litellm.ai · read 2 Oct 2026
- Integrations
- The integrations documentation lists observability tools including Langfuse, Datadog, Grafana Cloud, OpenTelemetry, LangSmith, Arize/Phoenix, and Helicone.docs.litellm.ai · read 2 Oct 2026
- Intended users
- The Enterprise plan is described for teams running LiteLLM in production, including high-volume and regulated deployments.litellm.ai · read 2 Oct 2026
- Product
- LiteLLM is an open-source AI gateway and LLM proxy for platform teams.litellm.ai · read 2 Oct 2026
- Routing
- LiteLLM supports load balancing, lowest-cost routing, and automatic routing of simple prompts to cheaper models.litellm.ai · read 2 Oct 2026
- Security
- The product page describes signed hardened non-root images, vulnerability scanning, SSO/JWT/RBAC controls, and secrets-manager integrations.litellm.ai · read 2 Oct 2026
- Spend controls
- It tracks usage and spend by key, user, team, organization, tool, agent, and MCP, and supports hard budgets and rate limits.litellm.ai · read 2 Oct 2026
- Support
- Enterprise includes dedicated support and onboarding, with 24/7 support and SLAs listed on the pricing page.litellm.ai · read 2 Oct 2026
- Trial
- The pricing FAQ offers an instant 30-day Enterprise trial key without a credit card.litellm.ai · read 2 Oct 2026
- Unified access
- The gateway provides one OpenAI-compatible API to 140+ providers and 1,800+ models.litellm.ai · read 2 Oct 2026
