z-node / docs

← Docs index

Aider

Setup guide for using Z-Node with Aider.

Official docs

Start with Aider's official OpenAI-compatible API and advanced model settings docs. They document the OPENAI_API_BASE override, openai/ model prefix, .aider.model.metadata.json, and .aider.model.settings.yml:

Z-Node example (.aider.conf.yml, .aider.model.metadata.json, .aider.model.settings.yml)

Copy this into your project root. It keeps your key in the environment, defaults to gpt-5.5, and hardcodes Aider's local model metadata/settings for Z-Node:

bash
cat > .aider.conf.yml <<'YAML'
model: openai/gpt-5.5
openai-api-base: https://api.z-node.com/v1
model-metadata-file: .aider.model.metadata.json
model-settings-file: .aider.model.settings.yml
YAML

cat > .aider.model.metadata.json <<'JSON'
{
  "openai/gpt-5.5": {
    "max_tokens": 128000,
    "max_input_tokens": 1050000,
    "max_output_tokens": 128000,
    "litellm_provider": "openai",
    "mode": "chat"
  },
  "openai/gpt-5.4": {
    "max_tokens": 128000,
    "max_input_tokens": 1050000,
    "max_output_tokens": 128000,
    "litellm_provider": "openai",
    "mode": "chat"
  },
  "openai/gpt-5.4-mini": {
    "max_tokens": 128000,
    "max_input_tokens": 400000,
    "max_output_tokens": 128000,
    "litellm_provider": "openai",
    "mode": "chat"
  },
  "openai/gpt-5.3-codex": {
    "max_tokens": 128000,
    "max_input_tokens": 400000,
    "max_output_tokens": 128000,
    "litellm_provider": "openai",
    "mode": "chat"
  },
  "openai/gpt-5.2": {
    "max_tokens": 128000,
    "max_input_tokens": 400000,
    "max_output_tokens": 128000,
    "litellm_provider": "openai",
    "mode": "chat"
  }
}
JSON

cat > .aider.model.settings.yml <<'YAML'
- name: openai/gpt-5.5
  edit_format: diff
  weak_model_name: openai/gpt-5.4-mini
  use_repo_map: true
  use_temperature: false
  accepts_settings: ["reasoning_effort"]

- name: openai/gpt-5.4
  edit_format: diff
  weak_model_name: openai/gpt-5.4-mini
  use_repo_map: true
  use_temperature: false
  accepts_settings: ["reasoning_effort"]

- name: openai/gpt-5.4-mini
  edit_format: diff
  weak_model_name: openai/gpt-5.4-mini
  use_repo_map: true
  use_temperature: false

- name: openai/gpt-5.3-codex
  edit_format: diff
  weak_model_name: openai/gpt-5.4-mini
  use_repo_map: true
  use_temperature: false
  accepts_settings: ["reasoning_effort"]
  overeager: true

- name: openai/gpt-5.2
  edit_format: diff
  weak_model_name: openai/gpt-5.4-mini
  use_repo_map: true
  use_temperature: false
  accepts_settings: ["reasoning_effort"]

- name: openai/codex-auto-review
  edit_format: diff
  weak_model_name: openai/gpt-5.4-mini
  use_repo_map: true
  use_temperature: false
YAML

Acceptable Z-Node model ids

Use these as openai/<model id> in Aider:

Smoke test

bash
OPENAI_API_KEY="$ZNODE_ISSUED_API_KEY" \
aider --message "Reply with exactly: Z-NODE-AIDER-OK" \
  --yes-always --no-git --no-auto-commits --no-dirty-commits --no-pretty