# RouterOne vs Vercel AI Gateway: no top-up fees, credit that never expires

> RouterOne and Vercel AI Gateway both bill at provider list prices with no markup. The differences: RouterOne charges no top-up fee, paid credit never expires and low-cost channels go far below official pricing while official channels are discounted too; it natively supports the Gemini protocol plus Alibaba Cloud Model Studio, Volcengine Seedance, MiniMax and xAI image and video protocols; and it provides enterprise-grade organization management with no account needed on another platform. Vercel AI Gateway is tightly integrated with the Vercel platform and the AI SDK, and offers BYOK, regional inference and enterprise SLAs.

Last verified 2026-10-19 · https://routerone.ai/vs/vercel-ai-gateway

- **0%** Top-up fee: Vercel AI Gateway passes payment processing fees on to you
- **Never** Paid credit expires: Vercel AI Gateway credits expire one year after purchase
- **4** Native text protocols: Including Gemini generateContent, which Vercel AI Gateway does not list

## RouterOne vs Vercel AI Gateway, feature by feature

| Feature | RouterOne | Vercel AI Gateway |
|---|---|---|
| Inference pricing | No inference markup; low-cost channels go far below official pricing, and stable official channels are discounted too | Billed at provider list price with no markup or platform fee; some models below list price, volume discounts negotiable |
| Top-up fee | 0% — credit received equals amount paid | Payment processing fees passed on to you; Enterprise can pay by invoice |
| Credit expiry | Paid credit does not expire | Purchased credits expire one year after purchase |
| Text protocols | Native OpenAI Chat, OpenAI Responses, Anthropic Messages and Gemini (generateContent), with most models callable across protocols | OpenAI Chat, Responses and Anthropic Messages; no native Gemini protocol listed in the docs |
| Image and video | Image: OpenAI Images and Alibaba Cloud Model Studio. Video: native Model Studio, Volcengine Seedance, MiniMax and xAI request formats plus a unified task API — existing client code only needs a new base URL and key | Images through an OpenAI-compatible API; video through an experimental AI SDK API |
| Embeddings and audio | Embeddings, speech and rerank endpoints (Coming soon) | Embeddings, text-to-speech, speech-to-text, realtime voice and rerank |
| Model coverage | 0+ text, image and video models, with more added regularly; 0 models can switch between providers | Advertises 400+ models |
| Routing and failover | Five routing strategies, configurable per API key; automatic failover on connection errors, timeouts, rate limits and server errors, with failed attempts not billed | Picks providers by uptime and latency; provider ordering, allowlists, sorting by cost or speed, and model fallbacks |
| Teams and budgets | Enterprise-grade organization management: organizations, workspaces and roles (including a finance role); spend limits per workspace, member and key on every organization — no enterprise plan required | Budgets per team, project, key and member on all plans; budgets are soft caps |
| Account requirement | Standalone account — sign up and start, no other platform needed | Requires a Vercel team account; apps do not have to be deployed on Vercel |
| Data privacy | Request and response bodies of successful text calls are not stored; customer content is never used for training | Prompt and response content not retained; zero data retention on Pro and Enterprise, billed per request when enabled team-wide |
| Usage and export | Itemized charges for every call; monthly statements exported to Excel by key, model and workspace | Logs exportable to CSV / JSON, up to 36 days back; per-request cost available via API |
| Single sign-on | SAML single sign-on (Coming soon) | SAML SSO as a Pro add-on or on the Enterprise plan |

## Choose RouterOne if

- You want no top-up fees, credit that never expires, and leading models far below official pricing through low-cost channels, or at a discount through stable official channels
- You use the native Gemini protocol, or want to call image and video models in the native Alibaba Cloud, Volcengine, MiniMax or xAI request formats
- You would rather not open a Vercel team account just for model access, and need your own organizations, workspaces and finance role
- You want failed retry attempts left off your bill

## Choose Vercel AI Gateway if

- You deploy on Vercel and use the AI SDK, and want keyless authentication and native integration
- You want to bring your own provider keys (BYOK) with no markup
- You need embeddings, speech or rerank endpoints today (on the RouterOne roadmap)
- You need pinned inference regions or a contractual SLA

## Switch from Vercel AI Gateway to RouterOne

Both work with the OpenAI and Anthropic SDKs and use the same manufacturer/model ID format. Switching only changes the endpoint and key — your application code stays as it is.

1. Sign up for RouterOne and create an API key in the console.
2. Replace the base URL with the RouterOne endpoint and the API key with your new key.
3. Keep your existing model IDs; for the few models whose version format differs, copy the ID from the model page.

- OpenAI SDK base URL: `https://gateway.routerone.ai/v1`
- Anthropic SDK / Claude Code base URL: `https://gateway.routerone.ai`

```python
import os
from openai import OpenAI

client = OpenAI(base_url="https://gateway.routerone.ai/v1", api_key=os.environ["ROUTERONE_API_KEY"])
response = client.chat.completions.create(
    model="anthropic/claude-sonnet-5",
    messages=[{"role": "user", "content": "Hello"}],
)
```

- With the AI SDK, switch to an OpenAI-compatible provider pointed at the RouterOne endpoint.
- Claude Code and the Anthropic SDK use the endpoint without /v1; the Gemini SDK can call generateContent directly.

## Frequently asked questions

### What is the main difference between RouterOne and Vercel AI Gateway?

Both bill at provider list prices with no markup. RouterOne charges no top-up fee, paid credit never expires, low-cost channels go far below official pricing while official channels are discounted too, and it natively supports the Gemini protocol plus vendor image and video protocols, and provides enterprise-grade organization management with no other platform account required. Vercel AI Gateway is tightly integrated with Vercel and the AI SDK, and offers embeddings and audio endpoints, BYOK, regional inference and enterprise SLAs.

### How can RouterOne price models below the official rate?

RouterOne connects each model to several providers in two kinds of channels: low-cost channels priced far below official pricing for cost-sensitive workloads, and official channels from model makers and major clouds that are stable and also discounted. Live per-provider prices are published on each model page; you can restrict an API key or workspace to official channels only, or set price-first routing.

### How do top-up costs compare?

On RouterOne you receive exactly what you pay, with no fee, and paid credit never expires. Vercel AI Gateway passes payment processing fees on to you (Enterprise can pay by invoice), and purchased credits expire after one year (as of 2026-10-19 — see the Vercel docs for current terms).

### What code changes are needed to switch from Vercel AI Gateway to RouterOne?

Usually two: point the base URL at the RouterOne endpoint and replace the API key with an RouterOne key. Model IDs use the same manufacturer/model format; check the RouterOne model page for the exact name of each model. AI SDK projects can switch to an OpenAI-compatible provider.

### Can I use RouterOne if my app is deployed on Vercel?

Yes. RouterOne is a standalone API service, independent of where your app runs. Add your RouterOne API key to your Vercel environment variables and call it with the OpenAI or Anthropic SDK.

### Can Claude Code, Cursor and similar tools switch to RouterOne?

Yes. RouterOne natively implements the Anthropic Messages and OpenAI protocols, and setup guides cover Claude Code, Codex, Gemini CLI, Cursor, Cline, Cherry Studio and more — replace the endpoint and key as described.

### Can I use Vercel AI Gateway and RouterOne together?

Yes. Both work with the OpenAI SDK, so you can split traffic by model or use case — for example, send mainstream models through RouterOne for lower prices and keep BYOK models on Vercel AI Gateway.

_Vercel AI Gateway and other product names mentioned on this page are trademarks of their respective owners. Information about third-party products is compiled from their official public sources for evaluation purposes only and may change; refer to each vendor's website for current details. To report an inaccuracy, contact support@amux.ai._