Junie Help

Model selection

Environment variable

CLI equivalent

Description

JUNIE_MODEL
--model

LLM model to use. If not specified, Junie uses the Default model — a dynamically set option with the best price-quality ratio.

JUNIE_LLM_PROVIDER
--provider

BYOK provider to use. Forces Junie to use a specific BYOK provider. Possible values: openai, anthropic, google, xai, openrouter, copilot, litellm. If not set, the Junie or Custom model provider is used.

JUNIE_EFFORT
--effort

Reasoning effort level for models that support it. Built-in levels: none, minimal, low, medium, high, xhigh, max. Supported values depend on the selected model and can include custom efforts defined in a custom model profile. Fails with an error listing the model's supported efforts if the value is unsupported or blank. Requires JUNIE_MODEL, --model, JUNIE_AGENT, or --agent.

To see all available models, run junie --help or use /model in interactive mode. To change the effort level during a session, use /model or the dedicated /effort command.

LLM providers

Junie supports four types of LLM providers:

  • Junie — models accessed through a JetBrains AI subscription, by means of a Junie login or a JetBrains AI API key. No additional configuration is required.

  • BYOK (Bring Your Own Key) — models accessed using your own API key from a third-party provider such as OpenAI, Anthropic, Google, xAI, OpenRouter, or GitHub Copilot. See BYOK for setup instructions.

  • Custom — models defined in a custom profile JSON file. See Custom LLM models for details, or Junie Local to run a model on your own machine.

  • Proxy — models accessed through a custom proxy endpoint configured in config.json.

When a model is available through multiple providers, Junie prefers the Junie provider by default. Use --provider to override this behavior.

The Junie provider's model list reflects confirmed JetBrains AI access. If the account is confirmed to have only Junie Lite access, only Junie Lite is offered from the Junie provider; BYOK, Custom, and Proxy providers keep their own model availability.

During standalone startup, model selection waits for the JetBrains entitlement check before automatically falling back to another provider. An explicit non-JetBrains provider selection does not wait.

During first-launch model onboarding, Skip selects a paid curated model only after standard JetBrains access is confirmed. Otherwise, it prefers Junie Lite when available; if Lite is not offered, it selects a model from the current catalog. This safe default does not hide models from the catalog.

An active AIF (AI Free) entitlement is not by itself treated as Junie Lite-only access. Premium model availability still depends on the standard entitlement result and backend authorization.

An explicit --model request for a premium JetBrains model fails with an access message when only Junie Lite is available. Specify --provider to use the same model through another available provider.

Provider auto-detection

When you specify --model without --provider, Junie resolves the provider automatically:

  1. If the Junie provider is available (you are logged in with a JetBrains account or API key), Junie uses it by default.

  2. If the Junie provider is not available (no login or API key), Junie looks through the connected BYOK providers and selects the first one that offers the requested model.

For example, running junie --model grok --grok-api-key <key> without a JetBrains login will automatically route to the xAI BYOK provider because it is the only available provider with the Grok model.

How Junie uses models internally

The model you select (or the Default) is your primary model — Junie uses it for the main reasoning and code generation work.

However, Junie also uses a separate model for certain internal tasks such as:

  • Summarizing context

  • Classifying and routing tasks

  • Extracting information for memory

  • Filtering capabilities

  • Other helper tasks

These internal tasks don't need the full power of the primary model, so Junie automatically picks an alternative from the same provider that is most suitable for these tasks.

This applies even when using BYOK. If you bring your own API key, Junie will still use a separate model from the same provider for these internal tasks. This means you may see API calls to a model you didn't explicitly select — that's expected and helps keep costs down and responses fast. For more details on configuring custom LLM providers, see Custom LLM Models.

Junie CLI also supports model aliases. Each alias maps to a specific model version set by the JetBrains team based on our own internal evaluations and is not necessarily the latest available version. The model behind each alias may change as new models are released.

Model ID

Provider

Current model

sonnet

Anthropic

Claude Sonnet 5.5

opus

Anthropic

Claude Opus 5.5

fable

Anthropic

Claude Fable 5.1

gpt

OpenAI

GPT-6.1 Sol

gpt-sol

OpenAI

GPT-6.1 Sol

gpt-codex

OpenAI

GPT-5.3 codex

gpt-astra

OpenAI

GPT-6 Astra

gpt-luna

OpenAI

GPT-6 Luna

gpt-terra

OpenAI

GPT-5.6 Terra

gemini-pro

Google

Gemini 3.1 Pro Preview

gemini-flash

Google

Gemini 3.8 Flash

grok

xAI

Grok 4.7

19 November 2026