# Onde Inference CLI: Fine-tune & Deploy Local LLMs

Onde Inference CLI is an MCP server for Claude Code that fine-tunes local Qwen models with LoRA and exports them as GGUF. Agents can also manage Onde apps and models.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 6 repo stars
- Author: ondeinference
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/ondeinference/onde-cli
- Updated: 2026-11-17
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/onde-inference-cli-fine-tune-deploy-local-llms

## What it does

Onde Inference CLI connects an agent to Onde Inference through an MCP server. It handles account access, apps, and the model catalog, while local Qwen fine-tuning uses LoRA and can produce GGUF exports.

The terminal client has `onde --tui` for sign-in and model management, plus `onde --mcp` for an stdio MCP connection. Claude Code starts it with `npx -y @ondeinference/cli`; an Onde Inference account is needed for sign-in.

## Requirements
- Needs: Node.js (commands run via npx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add onde-inference-cli-fine-tune-deploy-local-llms -- npx -y @ondeinference/cli
```

### Codex

**Platform CLI**
```bash
codex mcp add onde-inference-cli-fine-tune-deploy-local-llms -- npx -y @ondeinference/cli
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "onde-inference-cli-fine-tune-deploy-local-llms": {
            "command": "npx",
            "args": [
                "-y",
                "@ondeinference/cli"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "onde-inference-cli-fine-tune-deploy-local-llms": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "@ondeinference/cli"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add onde-inference-cli-fine-tune-deploy-local-llms npx -y @ondeinference/cli
```

## When to use it
- fine-tune a local Qwen model with LoRA
- export a tuned model as GGUF
- manage Onde apps from an agent
- inspect the Onde model catalog
- connect Onde Inference to Claude Code through MCP

## Example prompts
- Fine-tune the local Qwen model with LoRA for this project.
- Export the tuned model to GGUF.
- Show the apps and models available in Onde Inference.

## Similar
- [jetson-inference-mem-tune](https://affpapa.org/en/nevibecode/skills/jetson-inference-mem-tune.md) — skill · 1200 installs
- [llm-fine-tuning](https://affpapa.org/en/nevibecode/skills/llm-fine-tuning.md) — skill · 353 installs
- [serving-llms-on-epyc](https://affpapa.org/en/nevibecode/skills/serving-llms-on-epyc.md) — skill · 1300 installs
- [serving-llms-on-instinct](https://affpapa.org/en/nevibecode/skills/serving-llms-on-instinct.md) — skill · 1300 installs
- [gke-inference](https://affpapa.org/en/nevibecode/skills/gke-inference.md) — skill · 6791 installs
- [tao-run-inference-service](https://affpapa.org/en/nevibecode/skills/tao-run-inference-service.md) — skill · 1600 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
