The setup is one environment variable. The hard part is picking a local model large enough to drive an agent loop without stalling.

Why Would You Run a Coding Agent Against a Local Model?

Three reasons: source code confidentiality, no per-request cost, and no daily quota. Hospitals, defense contractors, and anyone under a strict data agreement usually cannot send repositories to a hosted API at all.

Can Qwen Code Use Ollama or vLLM?

Yes. Because it accepts any OpenAI-compatible endpoint, both work. Ollama exposes http://localhost:11434/v1 and vLLM serves the same shape on whatever port you launch it with.

Point the CLI at your server with the base URL and a placeholder key, then select the model name you loaded.

How Do You Configure It?

• Start your server: ollama serve or vllm serve Qwen/Qwen3-Coder-30B-A3B-Instruct
• Set OPENAI_BASE_URL=http://localhost:11434/v1
• Set OPENAI_API_KEY=local (any non-empty string)
• Set OPENAI_MODEL=qwen3-coder to match the loaded model
• Launch qwen and verify it responds

Which Local Model Should You Load?

Agentic coding needs tool-calling and long context. Below roughly 30 billion parameters, agents get confused about when to call tools and start hallucinating file contents.

• Best fit: Qwen3-Coder 30B-A3B — MoE, fast on a single 24GB GPU
• Quality first: Qwen3-Coder 480B if you have the VRAM or a multi-GPU box
• Avoid: 7B chat models; they patch the wrong files

What Breaks in Practice?

Two things. First, context length. Your local server may cap at 8K tokens while the agent assumes far more, so long files get truncated and edits land in the wrong place. Raise the server context window explicitly. Second, speed — a 480B model on one GPU can take thirty seconds per tool call, which makes an interactive loop painful.

Quick Summary

• Qwen Code works with any OpenAI-compatible endpoint, including Ollama and vLLM.
• Set the base URL, a dummy key, and the model name.
• Pick 30B or larger for reliable tool calling.
• Raise server context length or edits will truncate.
• Local models remove quota limits but cost GPU time.

That is twice the daily quota of its Gemini ancestor. The tradeoff is that the model behaves differently, and the ecosystem around it is younger.

What Is Qwen Code and How Is It Different From Gemini CLI?

Same interface, different brain. Qwen Code keeps the command structure and tool set of Gemini CLI but runs Qwen3-Coder, a 480-billion-parameter mixture-of-experts model with 35 billion active parameters tuned for agentic coding.

If you already know one, the other feels familiar within minutes.

How Do You Install Qwen Code?

• Install Node.js 20 or newer
• Run npm install -g @qwen-code/qwen-code
• Start it with qwen inside a project
• Choose Qwen OAuth and sign in through the browser
• Confirm the quota shows 2,000 requests

The OAuth step matters. Without it you land on an API key path and pay per token instead of using the free daily allowance.

Is Qwen Code Free?

Yes, 2,000 requests per day through Qwen OAuth, which is the largest free quota among the major CLI agents right now. No card required. The quota resets daily, and heavy agent loops can still exhaust it.

What Does the Agent Actually Do?

It edits multiple files in one turn and shows you a unified diff. It runs shell commands, greps the repo, and supports MCP clients so you can attach databases or internal APIs. The multi-file diff view is the part that saves the most time on real work.

Who Should Use It Instead of the Alternatives?

Choose Qwen Code if you want a big free quota and are comfortable with a fast-moving project. Look elsewhere if you need the maturity of Google’s backing, or if you are already standardized on a different model family.

What Are the Rough Edges?

Documentation lags the code, and version bumps sometimes change flag names. The model is strong at code but weaker than frontier models on ambiguous product questions. Give it precise instructions.

Quick Summary

• Qwen Code is Alibaba’s open-source fork of Gemini CLI on Qwen3-Coder.
• Install with npm install -g @qwen-code/qwen-code.
• Qwen OAuth unlocks 2,000 free requests per day.
• It edits several files per turn with a unified diff.
• Docs move slower than releases, so pin a version for team use.