01 / Routing
One endpoint for your tools
Use provider API credentials with scoped Router keys. Choose models, set limits, and configure explicit fallback chains for supported chat requests.
Model choice. API budgets. Codex capacity.
Run DeepSeek in Codex and connect compatible apps through one Router endpoint. Set daily budgets for API keys. See quotas and reset times across your connected Codex accounts.
Continue to ZergAI to check current account access. Existing users can sign in.
One endpoint / many routes
A Router key connects supported clients. Model availability depends on credentials, permissions, and protocol support.
curl https://zergrouter.com/v1/responses \
-H "Authorization: Bearer $ZERGROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "model": "modal/deepseek-v4.1-flash", "input": "Say hi" }'01 / Routing
Use provider API credentials with scoped Router keys. Choose models, set limits, and configure explicit fallback chains for supported chat requests.
02 / Subscriptions
Compare accounts by remaining weekly quotas, each quota’s reset date, and banked resets. Stale data stays clearly marked. API keys and personal subscriptions are separate access paths.
03 / Usage
Explore routed usage by model, date, and client. See input, cached input, and output where providers report them. Account-wide Codex history does not invent missing token breakdowns.
Available route / open-model inference
New personal accounts can try this model through Modal Shared Endpoints for 14 days after account creation, with a Router key, under a $0.50 UTC-day account-wide trial cap and the 5M-token monthly quota. Other Modal models are not included.
modal/deepseek-v4.1-flashAfter the trial, the paid route is metered at $0.30 / 1M input tokens, $0.03 / 1M cached input tokens, and $1.20 / 1M output tokens, up to your monthly usage cap. Trial access also depends on shared platform capacity and the Modal endpoint being available; check your authenticated catalog before connecting a client.
Try the DeepSeek quickstart →Works where you work
Connect a supported coding client or build with the Zerg ecosystem. Follow the setup guide for your tool.
Create your Zerg identity and use it to sign in across the stack.
Build and run autonomous software agents on the Zerg platform.
Work with agents from a fast, terminal-native coding environment.
See agent work, files, and execution state in a visual workspace.
Point the Codex Responses provider at one ZergRouter endpoint.
Bring your routed model catalog into the Pi coding agent.
Use ZergRouter as an OpenAI-compatible provider in OpenCode.
No. A Router API key authenticates routed requests; provider credentials and Codex subscriptions remain separate. New personal accounts have only the bounded Modal DeepSeek trial described above.
Yes. Use a model advertised by /v1/models?capability=codex_responses. The Modal route supports native Responses forwarding; the authenticated catalog is the authority for your available models. Follow the Codex setup guide.
Sign-up and Codex account monitoring are free. Eligible new personal accounts receive a bounded, 14-day DeepSeek trial on platform tokens. Beyond the trial, DeepSeek on our tokens requires a $5/month plan, metered usage, and a card on file. Your own provider keys are billed by your provider.
No. You choose the account for a local Codex session. Router shows quotas and reset times so you can decide which account to use.
Continue to ZergAI to check current account access. Existing users can sign in. Access and any billing requirements are confirmed in ZergAI, not inferred from this page.
Check access