> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://developers.meshapi.ai/docs/guides/llms.txt.
> For full documentation content, see https://developers.meshapi.ai/docs/guides/llms-full.txt.

# Dashboard Guide

> A guided tour of the Mesh API Control Panel.

The [Mesh Dashboard](https://app.meshapi.ai) is your central control center for managing AI models, keys, and billing. This guide walks through each section to help you make the most of the service.

## 1. API Keys

The **API Keys** section allows you to manage multiple sets of credentials. You can:

* **Create Keys**: Generate new tokens with optional labels.
* **Set Spend Caps**: Define a maximum USD limit for each key.
* **Monitor Usage**: See total requests, successful calls, and error rates per key.
* **Update Defaults**: Set a default model (e.g., `openai/gpt-3.5-turbo`) for when one is not specified in the request.

## 2. Billing

Mesh provides a unified balance across all providers.

* **Current Balance**: Real-time view of your remaining pre-paid credits.
* **Top-up**: Add credits via Stripe or local payment gateways.
* **Spend History**: Detailed breakdown of costs per provider and per model.
* **Auto-refill**: (Coming soon) Automatically refill your balance when it falls below a certain threshold.

## 3. Logs

The **Logs** section provides a historical record of every request made through your account.

* **Usage Events**: Filter logs by date range, model, or status (success/error).
* **Latency Tracking**: monitor the time taken by upstream providers.
* **Token Count**: precise breakdowns of prompt and completion tokens for every call.
* **Trace ID**: Unique IDs for every request to simplify debugging with the Mesh support team.

## 4. Models

Explore all available models and their current Pricing.

* **Discovery**: See which models are currently enabled (OpenAI, Anthropic, Gemini, Haiku, etc.).
* **Pricing**: live view of pricing per 1,000 tokens for both prompt and completion.
* **Free Models**: Easily filter for models with \$0 cost (ideal for testing).

## 5. Templates

Manage your prompts on the server to ensure consistency.

* **In-Dashboard Editor**: Write and test your system and user messages.
* **Parameter Tuning**: Set default temperature, max tokens, and stop sequences.
* **Variables**: Use `{{variable}}` syntax and pass values in the API request to dynamically render prompts.