OpenAI-compatible provider

Continue can use Tekiz as an OpenAI-compatible model provider with apiBase set to https://api.tekiz.ai/v1 or local port 8790.

Local Ollama fallback

Use local-first routes for cheap background edits, balanced routes for normal sessions, and stronger authorized cloud routes when local models are degraded or undersized.

429 failover

When a Continue route, provider key, or upstream model rate-limits, Tekiz can select another eligible route without editing every Continue config.

60-second Continue setup

{
  "models": [
    {
      "title": "Tekiz",
      "provider": "openai",
      "model": "auto",
      "apiBase": "https://api.tekiz.ai/v1",
      "apiKey": "sk_sage_your_generated_key"
    }
  ]
}

// Local router alternative:
// "apiBase": "http://localhost:8790/v1",
// "apiKey": "local-router"

Normal coding sessions

Use balanced routing for everyday Continue chat and edits, then escalate to stronger routes when quality or context needs rise.

Credential load balancing

Authorized credential pools can load balance and fail over on rate limits, provider health, latency, and policy.

Provider boundary

Tekiz does not grant unauthorized OpenAI, Anthropic, Gemini, NVIDIA, Ollama Cloud, or other provider access. Customers bring authorized credentials or use hosted routing for Tekiz account control.

Continue without provider-specific rewiring

Continue already supports OpenAI-compatible providers. Tekiz turns that single provider entry into a policy-managed model layer for local-first routing, failover, and observability.

Next steps

Create an API key, open the Continue guide, copy the hosted quickstart, or estimate routing fit.