> For the complete documentation index, see [llms.txt](https://docs.meganova.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.meganova.ai/faq/api-technical-specifications.md).

# API Technical Specifications

### Can I use MegaNova with Chub AI or Pygmalion-based frontends?

Yes. MegaNova is OpenAI-compatible, so you can use it with Chub AI, Pygmalion WebUI, or any frontend that supports custom API endpoints.

### How do I integrate MegaNova with SillyTavern or RisuAI?

Use endpoint `https://api.meganova.ai/v1/chat/completions`, API key (from dashboard, starts with `sk-`) and Model name (e.g., `meganova-ai/manta-mini-1.0`) to connect.

### Can I generate multiple API keys, and how does it work?

Yes, you can generate multiple API keys, but after generating a new one, the old key becomes unusable. Only 1 key works at a time. To switch, regenerate via the dashboard and update your config—keep backups securely.

> **Security Notice:** API key management is governed by our [Terms of Service](/legal-docs/terms-of-service.md). You are responsible for keeping your API keys secure and not sharing them with unauthorized parties.

### How do I manage multiple API keys securely?

Only 1 key active at a time (old keys deactivate after generating a new one). Store keys in a secure vault (e.g., env variables). Never share keys; regenerate immediately if compromised via dashboard.

### What API endpoints are available?

MegaNova supports the OpenAI API specification for both the `/completions` and `/chat/completions` endpoints, enabling seamless integration with any model using the same request/response format.

### What authentication methods are supported?

API keys (Bearer tokens) for chat/completions endpoints—generate via dashboard (starts with "sk-").

### What should I do for common errors like exceeding limits or connection issues?

**Exceed daily limit**: Switch to another model or reconfigure and reload the page. Sometimes the error may be caused by the platform—wait a few minutes and retry.

**No API access**: Some models require Engineer Tier 2 for free access; add a payment card (no charge) or deposit $10 to use them—no charges until you switch to paid versions.

**Free Models Available**: L3.3-MS-Nevoria-70B, L3-70B-Euryale-v2.1, L3-8B-Stheno-v3.2, DeepSeek-V3-0324, and Manta-Mini-1.0 are available for free without requiring payment card verification. These models have daily limits and are intended for evaluation and interactive use only.

**Unstable, No Response from Bot, or Provider return error**: Check API status at <https://status.meganova.ai/status/service> or contact support (always-on via Discord community). Try switching models (e.g., from Meta-Mini to Grok 4, Meta-Pro) or reducing context length. If it persists, share logs with the team—they offer 24/7 troubleshooting for prompts/scaling.

**Rate limited or connection issue**: Double-check your config for accuracy, or add a new config and reload the page.

**Exceed maximum context length**: Try a larger model like Manta-Pro, DeepSeek-R1 or reduce the context length for better results.

**Exceed current quota**: Check your balance and dashboard for billing status; switch to free models or deposit more to avoid this error.