OpenAI
Default: gpt-4o-mini
Opening Nythera
Opening Nythera
Nythera / Archive
Connect the model providers you already use. Nythera routes character generation and chat through your encrypted provider keys.
Important distinction
This manual explains how to connect OpenAI, Anthropic, Gemini, DeepSeek, xAI, and compatible endpoints. Nythera's internal application routes are not a supported public developer API.
Step 01
Open Settings → API Keys, select a provider, paste the key issued by that provider, and save it. The browser never receives the stored key again.
Open API KeysStep 02
OpenAI
Default: gpt-4o-mini
Anthropic
Default: claude-3-5-sonnet-latest
Gemini
Default: gemini-3.6-flash
OpenRouter
Default: openrouter/auto
DeepSeek
Default: deepseek-v4-flash
Mistral
Default: mistral-large-latest
Groq
Default: llama-3.3-70b-versatile
xAI (Grok)
Default: grok-4.3-latest
Step 03
Arrange saved providers, choose an exact model for each step, and move any provider into the primary position. Nythera advances past exhausted or invalid credentials and skips directly to the next provider during an outage; malformed requests still stop immediately.
Step 04
Use Add custom provider endpoint for an OpenAI-compatible service. Enter a unique provider ID, display name, base URL, default model, and API key. The endpoint must implement the OpenAI chat-completions request shape.
Base URL: https://your-provider.example/v1 Model: your-model-id Format: OpenAI compatible
Step 05