Which API should you use with Junie CLI?

For JetBrains Junie CLI on one monthly price, the best API is a flat-rate endpoint in a custom model profile: save ~/.junie/models/standardcompute.json pointing at https://api.stdcmpt.com/v1/chat/completions and run junie --model custom:standardcompute, from $19 a month with a $20 compute budget. JetBrains' own route is AI Pro at $10 or AI Ultimate at $30 a month, with 10 or 35 AI Credits worth $1 each. JetBrains says bring-your-own-key use is billed by the provider and needs no JetBrains AI subscription. Custom profiles are a Junie CLI feature; the Junie plugin inside the IDE does not take your own key.

Light or occasional use

If Junie fits inside AI Pro's 10 monthly credits, that is the cheapest paid route, and AI Free's 3 credits cover a light trial. A pay-as-you-go key suits occasional CLI runs.

One fixed monthly bill

When Junie CLI runs long tasks every day, a Standard Compute profile keeps the model bill at the plan price. Point fasterModel at an efficient model such as z-ai/glm-5.3-flash for helper tasks; requests stop at the monthly compute budget instead of billing overage.

API and subscription options for Junie

Prices and allowances can change. JetBrains AI prices and credits were read from JetBrains' licensing page, and the profile format and BYOK terms from Junie's docs, on October 14, 2026. JetBrains AI plans · Junie: custom LLM models · Junie: bring your own key. Spot an error? Tell us.

Create a Junie CLI model profile

{
  "id": "standardcompute",
  "displayName": "Standard Compute",
  "baseUrl": "https://api.stdcmpt.com/v1/chat/completions",
  "apiType": "OpenAICompletion",
  "apiKey": "${STANDARDCOMPUTE_API_KEY}",
  "fasterModel": { "id": "z-ai/glm-5.3-flash" }
}

Export STANDARDCOMPUTE_API_KEY and run junie --model custom:standardcompute, or pick it under the Custom tab of /model. The file name is the profile ID.

Junie questions

Why does my Junie custom model return a 404?

Junie uses baseUrl as the complete endpoint and appends no path. A baseUrl that stops at https://api.stdcmpt.com/v1 sends requests to /v1 itself, which returns 404; use https://api.stdcmpt.com/v1/chat/completions with apiType OpenAICompletion, or https://api.stdcmpt.com/v1/responses with OpenAIResponses. Also check the key variable: if a ${VAR} reference in apiKey is not set, the profile does not load.

Can I use my own OpenAI-compatible endpoint with Junie inside IntelliJ?

Not in the Junie plugin: JetBrains' docs say using your own API key with Junie is not supported in the plugin. Custom model profiles are a Junie CLI feature, so run junie in the IDE's terminal. For chat inside the IDE, JetBrains AI Assistant accepts OpenAI-compatible endpoints under Settings | Tools | AI Assistant | Providers & API keys.