Qwen Code with a local model: setup for Ollama, LM Studio or llama.cpp

Qwen Code is Alibaba's open-source (Apache-2.0) terminal coding agent, with VS Code, Zed and JetBrains integrations and an experimental web UI. Its docs show local servers such as Ollama and LM Studio configured as OpenAI-compatible providers in ~/.qwen/settings.json, which suits people already running Qwen-family models locally.

Source code Official docs Compare all agents

At a glance

Type: cli
Licence: Apache-2.0
Latest release: 0.25.3 (2026-10-19)
Status: Maintained
Local backends: ollama, vllm, openai-compatible
Needs native tool calling: Not verified
MCP support: Yes
Recommended minimum context: Not stated in its docs
Start-up prompt: Not measured
GitHub stars: 27.3k (2026-10-19)

How to set it up with a local model

Pick your backend. Each command has its own Copy button. Where a step uses an example model, swap in the one you run; the table below lists the ones that fit your machine.

Install it (macOS / Linux):

curl -fsSL https://qwen-code-assets.oss-cn-hangzhou.aliyuncs.com/installation/install-qwen-standalone.sh | bash

Start Ollama with a 64k context window (or set the context slider in the Ollama app's settings).
```
OLLAMA_CONTEXT_LENGTH=64000 ollama serve
```
In a second terminal, pull a tool-capable coding model (replace qwen3-coder:30b with any model you prefer).
```
ollama pull qwen3-coder:30b
```

Add Ollama as an OpenAI-compatible provider in your user settings (replace qwen3-coder:30b with the model you pulled; the key is a placeholder Ollama ignores).

~/.qwen/settings.json

{
  "env": {
    "OLLAMA_API_KEY": "ollama"
  },
  "modelProviders": {
    "openai": [
      {
        "id": "qwen3-coder:30b",
        "envKey": "OLLAMA_API_KEY",
        "baseUrl": "http://localhost:11434/v1",
        "generationConfig": {
          "contextWindowSize": 64000
        }
      }
    ]
  },
  "security": {
    "auth": {
      "selectedType": "openai"
    }
  },
  "model": {
    "name": "qwen3-coder:30b"
  }
}

Start Qwen Code in your project; switch models later with /model.
```
qwen
```

Other ways to install

npm (Node.js 22+)

npm install -g @qwen-code/qwen-code@latest

Homebrew

brew install qwen-code

Check that it works

Run a one-shot headless prompt. Inside an interactive session, /doctor shows the current authentication.

qwen -p "List the files in this directory and say which language the project uses."

Which local models fit, by machine

Checked by this site's engine at the context Qwen Code needs, with 32 GB of system RAM for MoE offload. A Mac counts three quarters of unified memory as usable.

Machine	Model	Quant	Context	Fit	Ollama
12 GB GPU	Qwen 3.5 9B	Q4_K_M	32,768	Comfortable (7.4 GB of 12 GB)	`qwen3.5:9b`
	Qwen 3.6 35B-A3B	Q4_K_M	32,768	With CPU offload (MoE) (22.7 GB of 12 GB)	`qwen3.6:35b-a3b`
16 GB GPU	Qwen 3.5 9B	Q4_K_M	32,768	Comfortable (7.4 GB of 16 GB)	`qwen3.5:9b`
	Qwen 3.6 35B-A3B	Q4_K_M	32,768	With CPU offload (MoE) (22.7 GB of 16 GB)	`qwen3.6:35b-a3b`
24 GB GPU	Qwen3.8 27B	Q4_K_M	32,768	Comfortable (19.9 GB of 24 GB)	`qwen3.8:27b`
	Qwen 3.6 27B	Q4_K_M	32,768	Comfortable (19.9 GB of 24 GB)	`qwen3.6:27b`
	Devstral Small 2 24B	Q4_K_M	32,768	Comfortable (20.7 GB of 24 GB)	`devstral-small-2:24b`
	Qwen 3.6 35B-A3B	Q4_K_M	65,536	Tight (23.3 GB of 24 GB)	`qwen3.6:35b-a3b`
32 GB GPU	Qwen3.8 27B	Q4_K_M	32,768	Comfortable (19.9 GB of 32 GB)	`qwen3.8:27b`
	Qwen 3.6 27B	Q4_K_M	32,768	Comfortable (19.9 GB of 32 GB)	`qwen3.6:27b`
	Devstral Small 2 24B	Q4_K_M	32,768	Comfortable (20.7 GB of 32 GB)	`devstral-small-2:24b`
	Qwen 3.6 35B-A3B	Q4_K_M	65,536	Comfortable (23.3 GB of 32 GB)	`qwen3.6:35b-a3b`
	Qwen3-Coder 30B-A3B (MoE)	Q4_K_M	65,536	Comfortable (25.7 GB of 32 GB)	`qwen3-coder:30b`
Mac, 64 GB	Qwen 3.6 35B-A3B	Q8_0	131,072	Comfortable (40.7 GB of 48 GB)	—

Your exact machine: filter the matrix or check any model in the analyzer.

Common problems with local models

contextWindowSize only tells Qwen Code the window size. Set the real window on the server: OLLAMA_CONTEXT_LENGTH for Ollama, --context-length in LM Studio, -c for llama-server.
Slow local servers can trip timeouts: the docs' local example raises generationConfig.timeout, and streamIdleTimeoutMs controls how long a model may stay silent between streamed chunks.
The settings file needs security.auth.selectedType = "openai" and model.name matching a provider id; without them you have to run /auth interactively.

Other coding agents

How they rank for local models: the best agent harness for local LLMs.

Sources (6), checked 2026-10-20