Zerg Router
Claim this toolUnified API routing for flexible AI model selection and cost control
Free, no paid tierVisit WebsiteFreeVisit Website
Tracked since2026
The Bottom Line
- Price
- Free, no paid tier. Compare plans
Key facts
- Acts as a unified API endpoint so you can use different AI models without changing your coding client
- Provides Codex subscription monitoring, API key budget controls, and usage analytics across models
- Offers platform-funded DeepSeek access on a pay-as-you-go basis, plus support for bringing your own provider credentials
Pros
- Preserves existing tooling workflow while adding model flexibility
- Granular budget controls and account monitoring help manage costs across teams
- No monthly base fee for platform-funded DeepSeek, pay only for what you use
Cons
- Platform-funded DeepSeek requires a card on file with no free trial; capacity is shared and may return 429 errors when exhausted
- Data residency guarantees are not provided for the shared DeepSeek route, not suitable for restricted data without a verified separate route
What is Zerg Router?
ZergRouter is an API routing layer that lets developers keep their existing coding tools while gaining flexibility in AI model selection. It provides a single OpenAI-compatible endpoint that can route requests to multiple models and providers, including platform-funded DeepSeek access, without requiring users to switch clients or manage separate credentials for each model.
The tool offers granular control over API usage: users can create scoped Router keys with optional daily budgets, monitor Codex subscription quotas and reset dates across connected accounts, and view detailed usage analytics by model, date, and client. It integrates with popular coding environments like Codex, Pi, OpenCode, and the Zerg ecosystem (ZTC, ZDE), enabling teams to standardize on one routing layer while preserving individual workflow preferences.
ZergRouter is designed for developers and teams who want to experiment with different models, enforce cost controls, and maintain visibility into AI usage, all without overhauling their existing toolchain. Platform-funded DeepSeek access requires a billing card, but Codex account monitoring and Router key creation are free.
Key Features
- Single OpenAI-compatible endpoint for routing to multiple models and providers
- Scoped Router API keys with optional daily budget limits per key
- Codex subscription quota monitoring with reset dates and freshness indicators
- Usage analytics broken down by model, date, and client (input/cached/output tokens)
- Platform-funded DeepSeek 4.1 Flash access through Modal or Fireworks (pay-as-you-go)
- Compatible with Codex, Pi, OpenCode, and the Zerg ecosystem (ZTC, ZDE)
- Explicit fallback chain configuration for supported chat requests
- Catalog-backed model availability checks via GET /v1/models endpoint
Pricing Plans
Pricing checked Sep 28, 2026
| Plan | Price | Details |
|---|---|---|
| DeepSeek 4.1 Flash (Metered) | $0.30 / 1M input tokens, $0.03 / 1M cached input tokens, and $1.20 / 1M output tokens |
|
Is Zerg Router worth the price?
Good value
This metered tier is genuinely cheap at $0.30 per million input tokens and $1.20 per million output tokens, undercutting most proprietary APIs, though the 5M-token monthly router quota may force upgrades for heavier users.
The free tier adds real value for testing. Best for cost-conscious developers prototyping small workloads.
Hidden Costs & Gotchas
5M-token monthly Router quota
Reviews
$99Free with your review
Write a reviewReview Zerg Router, get a free AI guide
Share your experience and we will send you Improve Your Thinking Patterns Using ChatGPT, free.
Best Zerg Router Alternatives
Top alternatives based on features, pricing, and user needs.
GoModelPaid
Unified AI gateway for routing, caching, and controlling requests
Anthropic APIPaid
API access to Claude for building AI applications
OpenAI APIPaid
Programmatic access to OpenAI's full model lineup: GPT-5.5, o3, gpt-image-1
OllamaFree
Run open-source LLMs locally with one command
LiteLLMFreemium
An AI Gateway for unified access, cost tracking, and fallbacks across 100+ Large Language Models.
LM StudioFree
Run local LLMs with a beautiful interface
Still deciding?
Most buyers shortlist 2 or 3 tools before committing. Pull a side-by-side comparison or browse the full alternatives shortlist below.
Explore More
Zerg Router FAQ
How does Router handle my existing Codex subscription?
Router does not switch your local Codex account automatically. The Subscriptions feature lets you monitor a connected personal Codex account's weekly quota, reset timing, and banked resets, but you choose which account to use for a session. Router routing and platform-funded DeepSeek do not spend your ChatGPT or Codex subscription quota.
Can I use my own provider API keys for models other than DeepSeek?
Yes. You can bring provider API keys and use them with scoped Router keys. The authenticated model catalog determines what models are available based on your credentials, permissions, and protocol support. Other routes require your own provider credential or an explicit funding arrangement.
What happens if the platform-funded DeepSeek capacity is exhausted?
If global shared capacity is exhausted, the request returns a 429 HTTP status. There is no funded trial, and without paid access the request is refused with 402. You should check your authenticated catalog before configuring a client to ensure the route is available.
Does creating a Router key automatically grant model access?
No. A Router key authenticates and scopes requests but does not itself purchase a subscription. The route must have an eligible credential (your own or platform-funded) and you must set up billing if using platform-funded DeepSeek. Check the route and funding path before running paid work.
Can I run my own model deployment through ZergRouter?
The shared platform-funded route does not create or control your own deployment. If you need a specific deployment location or operator control, you must use a separately configured route whose serving path and terms you have verified. ZergRouter supports custom routes if you bring your own provider credentials.
How are DeepSeek requests routed and where does data go?
Your client sends the request to ZergRouter, which forwards the selected model request to Modal Shared Endpoints or Fireworks for DeepSeek inference. This is a shared route, not a customer-controlled deployment. ZergRouter does not claim a specific data-residency region; review the route and provider terms before sending restricted data.
What tools and clients are compatible with ZergRouter?
ZergRouter works with any OpenAI-compatible client. Officially supported coding tools include Codex (via the Responses provider), Pi, OpenCode, and the Zerg ecosystem—ZergAI (for identity), Zerg (autonomous agents), ZTC (terminal-native coding), and ZDE (visual workspace). The Router endpoint is language-agnostic and can be called via cURL, Python, or TypeScript.
Is there a way to set a hard spending cap on paid routes?
Yes. In the dashboard you can set optional daily budgets per Router key. Additionally, a monthly usage cap applies to platform-funded DeepSeek access. The 5M-token monthly Router quota and your usage cap are separate controls, and you can monitor usage to stay within limits.
Source: zergrouter.com