Organization repository on GitHub
💭 Free AI and image generation for teens
- Forks
- 54
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Repos & SDKs
Hack Club AI
A lightweight, self-hosted LLM proxy that gives Hack Club's teen community free chat, embedding, and image models over OpenAI-compatible endpoints, with Hack Club auth, API keys, moderation, spending limits, and analytics.
- Stars
- 136
- Language
- TypeScript
- Reading time
- 2 min
- Published by
- Community
- Author
- hackclub
- Last verified
- Oct 2026
- Languages
- TypeScript
- Platforms
- APIPostgreSQLOpenRouter
- Capabilities
- Open sourceSelf-hostedRequires API key
Highlights
- Implements Hack Club auth, API keys, chat and embedding model support, moderation model support, global analytics, and usage logging.
- Must sit behind a reverse proxy such as Traefik so IPs cannot be spoofed; the README also recommends OpenRouter as the main provider.
- Model allowlists are env-configured: ALLOWED_LANGUAGE_MODELS, ALLOWED_EMBEDDING_MODELS, and ALLOWED_IMAGE_MODELS.
- Stack: Bun runtime, Hono server, Postgres 18 with Drizzle, Alpine plus htmx and hono/jsx, Biome for formatting and linting.
- Optional features: Sentry error tracking, Replicate for image generation and speech, PostHog analytics, and coding-tool blocking for Copilot and Cursor.
Watch out
Copy reviewed against the repo README on 2026-10-15. The README documents no Jev or TypeSafe integration; this is a general-purpose LLM proxy listed in the logicrw collection. The digest metadata shows no license, and self-hosting needs Hack Club OAuth credentials, provider keys, a Postgres database, and a reverse proxy.
Questions about this entry
Short answers drawn from the linked project or page.
A free AI service for teens, run as a lightweight LLM proxy: chat, embedding, moderation, and image generation behind OpenAI-compatible endpoints with Hack Club authentication, API keys, usage logging, and spending limits.
The README documents no Jev or TypeSafe integration. It fronts providers like OpenRouter and optionally Replicate and OpenAI's moderation API, and it appears in this catalog because it is listed in the logicrw collection, not because it calls Jev.
Allowlists set in environment variables: language models such as qwen3-32b, kimi-k2-thinking, gpt-oss-120b, gemini-2.5-flash and deepseek-v3.2-exp; embedding models like qwen3-embedding-8b; and image models like gemini-2.5-flash-image.
The Bun runtime, a Postgres 18 database, Hack Club OAuth client ID and secret, upstream provider keys, and a reverse proxy such as Traefik in front so IPs cannot be spoofed. Some environment variables are not needed when using Docker Compose.
The GitHub description says free AI and image generation for teens, but running your own instance means you pay the upstream providers. The README notes that OpenAI's moderation model is free but needs an active balance, and it recommends OpenRouter for easier billing.
Bun as the runtime, Hono for the server, Postgres with Drizzle as the ORM, Alpine with htmx and hono/jsx for the frontend, OpenRouter as the main provider, Replicate for media, PostHog and Sentry optional, and Biome for formatting and linting.
It is a proxy, not a model or an SDK. The README says the code is probably not the best but it works, and it must run behind a reverse proxy to prevent IP spoofing before it is safe to serve users; coding tools such as Copilot and Cursor are blocked.
More Jev resources like this
A host-agnostic routing layer that sits between a coding harness and its models, choosing capacity per round from trajectory evidence like tool calls, failures and context pressure, running as a user-scoped daemon and local OpenAI-compatible proxy.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
vizuh/sabiTypeScript
vizuh/sabiTypeScriptRun Laya locally on Bun with native ONNX inference and Hugging Face tokenizers: an HTTP server implementing TypeSafe's decision API plus an English/Persian playground with 16 demos, usable through the official @typesafe-ai/sdk with a custom baseURL.
Jev Repos & SDKs#community#catalog-import#logicrw
Community
ali-master/usejevTypeScript
ali-master/usejevTypeScriptTyped semantic decisions for TypeScript: ask Zod 4 or native question batches about your data and get validated answers through async/await or Web Streams, with confidence gates, retries, replay and observability. Jev runs via TypeSafe AI or Vercel evaluation.
Jev Repos & SDKs#community#catalog-import#logicrw
Community
nitoba/questionsTypeScript
nitoba/questionsTypeScriptReqLLM is an Apache-2.0 Elixir library built on Req and Finch that standardizes LLM requests and responses across providers, exposing 1,205 models across 21 integrations behind one unified API.
Jev Repos & SDKs#community#catalog-import#logicrw
Community
agentjido/req_llmElixir
agentjido/req_llmElixirA small, idiomatic Elixir client for Jev: Jev.ask?/2 returns a real boolean, choose/3 picks among supplied options, score/3 rates ordered levels, and analyze/3 sends several typed questions in one provider request.
Jev Repos & SDKs#community#catalog-import#madewithjev
Community
elue-dev/jev_elixirElixir
elue-dev/jev_elixirElixirTurns any Hugging Face causal LM plus a few thousand labelled decisions into a Jev-style model that answers Choice, Score and Noul questions in one forward pass, with calibrated probabilities and zero generated tokens.
Jev Repos & SDKs#community#catalog-import#madewithjev
Community
hwfengcs/any2jevPython
hwfengcs/any2jevPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.