README.md

---
title: MCP Latency Estimator
emoji: ⚡
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: "4.44.0"
app_file: app.py
pinned: false
license: apache-2.0
tags:
  - mcp
  - latency
  - mobile
  - benchmark
  - dispatchai
---

# ⚡ dispatchAI MCP Latency Estimator

**MCP Server** — estimate on-device inference latency for any mobile LLM.

## Tools

- `estimate_latency(params, quant, hardware, prompt_tokens, generate_tokens)` — Get tokens/sec, RAM, and total time
- `list_supported_hardware()` — Browse supported hardware profiles

## Usage in Claude Desktop

```json
{
  "mcpServers": {
    "dispatchai-latency": {
      "url": "https://huggingface.co/spaces/dispatchAI/mcp-latency-estimator/mcp"
    }
  }
}
```

🚀 [dispatchAI](https://huggingface.co/dispatchAI) — Small. Mobile. Free. UAE-built.