# Alternatives to LiteLLM

> Source: https://theagentsindex.com/litellm/alternatives/ (re-verified)
> Updated: 1 November 2026

## At a glance

| Name | Best for | Starts at | Better than LiteLLM at | Checked |
| --- | --- | --- | --- | --- |
| [LiteLLM](https://theagentsindex.com/litellm.md) | Platform teams operating production AI across models, agents and MCP tools | Free | — | 29 October 2026 |
| [OpenRouter](https://theagentsindex.com/openrouter.md) | Developers building model-flexible AI products across providers | Free | How it meters | 29 October 2026 |
| [Portkey](https://theagentsindex.com/portkey.md) | AI teams running production applications across multiple LLM providers | Free, then $49 a month | How it meters | 29 October 2026 |

The ones people open most here come first, then the closest matches.

## Shortlist

### 1. OpenRouter

One OpenAI-compatible API that routes across hundreds of models and providers.

**Where it improves on LiteLLM:**
- [How it meters: yes — Inference is usage-metered per input and output token at each provider's posted model rate.](https://openrouter.ai/pricing)

**Trade-offs against LiteLLM:**
- [Data residency: partial — Business and Enterprise can route prompts and completions within Europe or the United States, while account records may remain outside that boundary.](https://openrouter.ai/docs/guides/features/in-region-routing)

[View OpenRouter](https://theagentsindex.com/openrouter)

### 2. Portkey

An open-source AI gateway that unifies model access with governance and LLM observability.

**Where it improves on LiteLLM:**
- [How it meters: yes — The paid Production plan bills monthly, with a stated request-overage charge after its included recorded-log allowance.](https://portkey.ai/pricing)

**Trade-offs against LiteLLM:**
- [Runs in your cloud: partial — Enterprise hybrid deployment runs the AI Gateway and data plane in the buyer's environment while Portkey hosts control management.](https://portkey.ai/docs/product/product-feature-comparison.md)
- [Admin controls: partial — Enterprise administrators can control model access, workspace budgets, rate limits, and permissions, but these controls are not product-wide.](https://portkey.ai/docs/product/enterprise-offering/security-portkey.md)
- [Data residency: partial — Hybrid Enterprise deployment keeps sensitive LLM data in the buyer's infrastructure, though no specific regional jurisdiction is published.](https://portkey.ai/docs/product/product-feature-comparison.md)

[View Portkey](https://theagentsindex.com/portkey)