# Ephemeris Time-Series Forecasting

> Ephemeris Time-Series Forecasting is an MCP server for analytics and data. MCP server for probabilistic time-series forecasts from zero-shot foundation models. It can run locally on your machine or as a hosted remote server.

- **Kind:** MCP server
- **Pricing:** Free trial
- **Platforms:** API, CLI, Web
- **Website:** https://ephemeris.cascade.industries
- **Repository:** https://github.com/tensorlink-ai/ephemeris-mcp
- **MCP Registry name:** industries.cascade/ephemeris
- **Categories:** analytics and data, AI models and LLM tooling
- **Listing updated:** 2026-11-15

## About

Ephemeris gives an agent access to a panel of open-weights forecasting models, listed on the site as seven: Chronos-2 (Amazon), TimesFM 2.5 (Google), Toto 2 (Datadog), TiRex-2 (NXAI), PatchTST-FM r2 and FlowState r1 (IBM Research) and Timestral 209M (Mistral AI). An agent sends history and gets quantile forecasts back, with no training or feature engineering.

A forecast can name one model, let Ephemeris route to a fit for the data, or use an ensemble that blends six of the seven and costs a flat 5 credits per series per 64 steps. The service reports benchmark scores for the ensemble on TIME and GIFT-Eval, using each benchmark's own harness, and notes a few leaderboard entries score better on GIFT-Eval. Saved forecasts can be described or plotted without running inference again.

Pricing is prepaid credits with no subscription: one model run is 1 credit, about $0.001, and new accounts get 2,000 free. Access needs an API key, created through OAuth sign-in or the npx ephemeris-mcp login command.

## Features

- **Seven models:** Panel of open-weights time-series foundation models.
- **Route and ensemble modes:** Pick a model automatically or blend several.
- **Quantile forecasts:** Returns prediction intervals, with up to 64 series per call.
- **Plots and summaries:** Describe or plot a saved forecast without new inference.
- **Usage tracking:** Every response reports credits charged and balance.

## Good for

- Forecast weekly sales or demand
- Project cash flow with a range of outcomes
- Predict web traffic or energy load
- Sanity-check an agent's numeric estimate against a model

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http ephemeris https://ephemeris.cascade.industries/api/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://ephemeris.cascade.industries/api/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "ephemeris": {
      "url": "https://ephemeris.cascade.industries/api/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "ephemeris": {
      "type": "http",
      "url": "https://ephemeris.cascade.industries/api/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "ephemeris": {
      "serverUrl": "https://ephemeris.cascade.industries/api/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "ephemeris": {
      "type": "http",
      "url": "https://ephemeris.cascade.industries/api/mcp"
    }
  }
}
```

## Tools

- `get_version`: Hosted server version and supported artifact schemas
- `forecast`: Forecasts 1 to 64 series in route, ensemble or explicit mode
- `describe_forecast`: Interprets a saved forecast artifact without new inference
- `plot_forecast`: Renders a saved forecast as PNG or SVG
- `list_models`: Lists the live model panel with health, limits and prices
- `get_balance`: Returns spendable credits
- `get_usage`: Lists recent requests and their costs

## FAQ

### How is it priced?

Prepaid credits, with one model run costing 1 credit (about $0.001) and an ensemble 5 credits; packs start at $10 for 10,000.

### Is there a free allowance?

New accounts get 2,000 free credits.

### How do I authenticate?

Use an API key (pc_live_...) as a bearer token or sign in with OAuth.

### What tools does it expose?

forecast, describe_forecast, plot_forecast, list_models, get_balance, get_usage and get_version.

### Is Ephemeris Time-Series Forecasting free?

Ephemeris Time-Series Forecasting is a paid product with a free trial.

### How do I add Ephemeris Time-Series Forecasting to Claude Code?

Run this command in your terminal: claude mcp add --transport http ephemeris https://ephemeris.cascade.industries/api/mcp

### What are the best alternatives to Ephemeris Time-Series Forecasting?

The closest alternatives to Ephemeris Time-Series Forecasting are Agent-Native Analytics, Claude Code Ultimate Guide, Google Ads + Meta Ads + SEO/GEO MCP, Tapetide - Indian Stock Market MCP and World Monitor.

## Alternatives

- [Agent-Native Analytics](https://ai-powered.com/mcp/agent-native-analytics): Agent-Native Analytics is an open-source MCP server for analytics and data.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Google Ads + Meta Ads + SEO/GEO MCP](https://ai-powered.com/mcp/google-ads-meta-ads-seo-geo-mcp): Google Ads + Meta Ads + SEO/GEO MCP is a freemium MCP server for marketing and seo.
- [Tapetide - Indian Stock Market MCP](https://ai-powered.com/mcp/tapetide-indian-stock-market-mcp): Tapetide - Indian Stock Market MCP is a freemium MCP server for finance and payments.
- [World Monitor](https://ai-powered.com/mcp/world-monitor): World Monitor is a freemium MCP server for analytics and data.
- [CompanyData](https://ai-powered.com/mcp/companydata): CompanyData is a paid MCP server for analytics and data.
- [ChartAnt](https://ai-powered.com/mcp/chartant): ChartAnt is a freemium MCP server for design and media.
- [Quantivex Research](https://ai-powered.com/mcp/quantivex-research): Quantivex Research is a free MCP server for finance and payments.

---
Source: https://ai-powered.com/mcp/ephemeris-time-series-forecasting
