Jev LLM

An experimental semantic decision runtime for ordinary Hugging Face causal language models: it returns typed Boolean or Choice distributions straight from logits, reuses context prefill across fields and can compile repeated fields into small classifiers.

ReqLLM 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.

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.

A small Python support-ticket triage pipeline: Jev judges every incoming ticket in one typed call, plain if statements decide whether it is worth a second pass, and an LLM writes a reply only when Jev's scores cross a threshold.

A BYOK workbench for practicing Jev concepts without official access: emulates State, Noul, Score and Choice via Groq, OpenAI, Anthropic, Gemini or local Ollama, or runs against native Jev with authorized credentials.

LLM-as-Jev asks how far off-the-shelf LLMs already go as Jev-style decision models and when fine-tuning is needed. Frozen Qwen3.5-4B and Qwen3-0.6B match or exceed existing training-free readouts on the same 4B backbone, and Thinking Jev adds budgeted test-time reasoning.

A semantic cache for LLM queries, fully integrated with LangChain and llama_index. Zilliz's cache integrates TypeSafe Jev Noul checks to evaluate cache hit freshness and time-dependent query validity, cutting repeated requests and API cost.