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

Aider is an open-source (Apache-2.0) AI pair-programming tool for the terminal that edits files with text edit formats rather than native tool calls. It works with Ollama, LM Studio and any OpenAI-compatible server; development has slowed, with the last PyPI release (0.86.2) on 2026-02-12.

Source code Official docs Compare all agents

At a glance

Type: cli
Licence: Apache-2.0
Latest release: 0.86.2 (2026-02-12)
Status: No release since 2026-02-12
Local backends: ollama, openai-compatible
Needs native tool calling: No
MCP support: Not verified
Recommended minimum context: Not stated in its docs
Start-up prompt: Not measured
GitHub stars: 49.6k (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 -LsSf https://aider.chat/install.sh | sh

Point aider at your Ollama server.

export OLLAMA_API_BASE=http://127.0.0.1:11434

Pull a model (replace qwen3-coder:30b with any model you prefer).
```
ollama pull qwen3-coder:30b
```
Start Ollama; aider's docs use an 8k default context here because aider sizes num_ctx per request itself.
```
OLLAMA_CONTEXT_LENGTH=8192 ollama serve
```
In another terminal, run aider from your project with the ollama_chat/ prefix.
```
aider --model ollama_chat/qwen3-coder:30b
```
Optional: pin a fixed context window instead of aider's per-request sizing.
.aider.model.settings.yml
```
- name: ollama_chat/qwen3-coder:30b
  extra_params:
    num_ctx: 65536
```

Other ways to install

pip

python -m pip install aider-install && aider-install

uv tool install --force --python python3.12 --with pip aider-chat@latest

Check that it works

Start aider in a git repository and ask it to describe one file. With Ollama, `ollama ps` shows the context aider requested.

ollama ps

Which local models fit, by machine

Checked by this site's engine at the context Aider 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

Use the ollama_chat/ prefix; aider's docs recommend it over ollama/.
Aider's docs still say Ollama defaults to a 2k window; current Ollama defaults to 4k under 24 GiB VRAM. Aider sets num_ctx per request either way, so data isn't silently dropped.
LM Studio needs LM_STUDIO_API_KEY set to a dummy value, or requests fail with an empty Bearer token.
Models aider doesn't know trigger model warnings; aider's docs cover what they mean.

Other coding agents

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

Sources (8), checked 2026-10-20