[heydryft]

Adds an AI Endpoints guide for Runcrate, an OpenAI-compatible API for open models (DeepSeek, GLM, Qwen, gpt-oss, Kimi). It follows the NEAR AI Cloud page: .env key, an endpoints.custom entry with baseURL: https://api.runcrate.ai/v1, fetch: true, and notes on model IDs.

• content/docs/configuration/librechat_yaml/ai_endpoints/runcrate.mdx (English source; translations left to the translation workflow)
• meta.json page entry, locale sidebars synced with pnpm sync:locale-meta
• components/icons/providers/runcrate.tsx (currentColor) registered in index.ts and lib/icons.tsx

Checks: pnpm check:locale-meta passes, pnpm typecheck clean, Prettier clean on the changed files, pnpm test 39 files / 445 tests pass. The config was checked against the live API: GET https://api.runcrate.ai/v1/models is public and chat completions with deepseek-ai/DeepSeek-V4.1-Flash return 200.

Disclosure: I work on Runcrate.

[vercel]

@heydryft is attempting to deploy a commit to the LibreChat's projects Team on Vercel.

A member of the Team first needs to authorize it.