LiteLLM vs Vercel AI Gateway
Both are alternatives to OpenRouter. Here's how they stack up — verified facts, no spin.
Also searched as Vercel AI Gateway vs LiteLLM — same comparison, one verdict.
LiteLLM is open source (MIT (core); enterprise/ directory under a separate commercial licence) and Vercel AI Gateway is not (Proprietary (hosted service)) — so the real question is whether you want to own the llm gateways & routing stack or rent it.
LiteLLM
TOP PICKThe open gateway most teams run: 100+ providers, your keys, no fee on spend.
A proxy with an OpenAI-compatible endpoint in front of 100+ providers, with virtual keys, per-team budgets and fallbacks. You run it and pay the providers directly, so nothing is added to your token spend. GitHub showed 63,375 stars and a push on 17 Nov 2026. The core is MIT; the enterprise/ directory is under a separate commercial licence.
Vercel AI Gateway
A hosted gateway with no markup on tokens.
Vercel's hosted gateway: one endpoint and one bill across many providers, with no markup and no platform fee on tokens by its pricing docs (updated 8 Sep 2026). The free tier covers a subset of models with rate limits; on paid use you can bring your own keys at zero markup, and you pay the payment-processing fees. Some add-ons cost extra, among them a team-wide model allowlist, regional hosting at 5%, and bring-your-own-key without fallback.
Side by side
10 points of comparison, every one read from a verified field. Green marks the side that wins a row outright. A dash means we do not hold that fact — never that it is zero.
| LiteLLM | Vercel AI Gateway | |
|---|---|---|
| Sovereignty ScoreOur transparent 0–100 composite for data ownership and exit cost. | 90 | 30 |
| Open source | Yes | No |
| Self-hostable | Yes | No |
| Local-first data | Yes | No |
| License | MIT (core); enterprise/ directory under a separate commercial licence | Proprietary (hosted service) |
| Pricing | Free and self-hosted; you pay the model providers directly. Enterprise features such as SSO and audit logs are sold separately. | No markup and no platform fee on tokens; payment-processing fees are passed on. Free tier with a subset of models and rate limits. Add-ons such as regional hosting (5%) cost extra. |
| RAM to run it wellThe figure that actually matters, not the vendor's minimum. | 2 GB | — |
| Realistic running costWhat the box costs each month if you run it yourself. | $12–24/mo, against LLM gateway products charging per request | — |
| Setup timeHonest first-install estimate, not the marketing quickstart. | 1 hour | — |
| Ongoing maintenanceThe part nobody budgets for. | Low, but it is now in the path of every model call you make. | — |
LiteLLM is Macrostack's recommended OpenRouter alternative, so it's our pick here.
Weighing both against staying on OpenRouter? Is OpenRouter free? What it actually costs →
LiteLLM
Strengths
- +No fee on token spend: providers bill you at their own price
- +Prompts never leave your infrastructure
- +Virtual keys, per-team budgets and fallbacks built in
Trade-offs
- −You run it: deployment, upgrades and uptime are yours
- −SSO and audit logs sit behind the commercial licence
Vercel AI Gateway
Strengths
- +No markup on tokens
- +Nothing to run
- +Bring your own keys at zero markup on paid use
Trade-offs
- −Hosted and proprietary: every prompt passes through Vercel
- −Add-ons such as regional hosting carry a surcharge
Which one fits you
The trade-offs above, turned into a decision. Find the line that describes your team.
Choose LiteLLM
if you want the source and the option to fork it, and you would rather run it on your own hardware, and your data should stay on machines you control.
Choose Vercel AI Gateway
if no markup on tokens.
Neither, yet
if both carry a real cost you should weigh first — you run it: deployment, upgrades and uptime are yours, and hosted and proprietary: every prompt passes through Vercel. If either of those is a dealbreaker for your team, the shortlist is wrong rather than the choice.
What it takes to run these yourself
Real requirements and honest running costs, not the vendor quickstart.
LiteLLM vs Vercel AI Gateway — common questions
Is LiteLLM a better fit than Vercel AI Gateway for llm gateways & routing?
It depends on what you are optimising for, and the honest split is this: LiteLLM scores 90 to Vercel AI Gateway's 30 on data ownership and exit cost, so it is the safer choice if you care about being able to leave. Vercel AI Gateway earns its place on a different axis — no markup on tokens. Neither is a wrong answer for every team; the table above is the actual comparison.
What happens if we want to switch later?
LiteLLM keeps its data local or in open formats, so leaving is an export rather than a negotiation. Vercel AI Gateway keeps your data on its own infrastructure, which is the part to plan for: confirm the export path before you commit, not when you are already trying to leave.
Can I self-host LiteLLM?
LiteLLM can be self-hosted; Vercel AI Gateway cannot, so it stays on the vendor's infrastructure whatever you do.
Are LiteLLM and Vercel AI Gateway both alternatives to OpenRouter?
Yes — both appear in our OpenRouter comparison, which is why they are worth putting side by side. People usually arrive here already having decided to move off OpenRouter and now choosing between the two replacements, which is a narrower and much easier question.
More llm gateways & routing comparisons
Related alternative guides
Facts verified 2026-11-17. Licenses and pricing change — spotted something out of date? That's a correction we want.