Search

LLM inference runtime for OwlMeans applications: a resilient four-method model over any LangChain chat model, a provider-plugin layer, a model factory service, and an execution abstraction that resolves models from an inheritable policy.

A lightweight, CPU-optimized RAG library powered by a local Small Language Model (SLM)

Single client-side AI client for the fleet — wraps g4f (@gpt4free/g4f.dev) with multi-provider failover (default auto-router → DeepInfra → Puter). Framework-agnostic TS. chat/complete/vision/image/listModels, streaming, AbortSignal. Change the provider st

LangChain Messages provider for @launchdarkly/ai-server

LiteLLM custom provider for Databricks Model Serving with live fuzzy model resolution

Jupyterlab extension to replace individual jupyterlab_*_code_extension and integrate into one, that offers joint capability, joint settings page and individual right toolbar panels per code assistant, with the ability in settings to decide which code assistant support is on (default - all are enabled)

React Native library for on-device audio AI pipelines (VAD, turn detection, ONNX) powered by the Switchboard SDK.

Installer for the kijito-claude toolkit: copies the context-tracking, session catch-up, and self-clear scripts plus the Kijito skills into ~/.claude.

A modular, multi-client library for aiogram to handle user response waiting in Telegram bots.

AI-security auditor — score any AI endpoint against Responsible-AI metrics, plus a 44-control safety register (early/experimental)

The embedding lifecycle, end to end — recommend, generate, store, search, cache, track, model, inspect, and visualize embeddings behind one consistent surface.

Canonical SQLite storage engine, DDL migrations, Drizzle ORM, JCS hash, and graph query engine for PAKB

Experimental OpenCode web-search provider backed by a user's ChatGPT OAuth connection.

AgentRAG SDK — x402-paid retrieval over your own documents, for agents.

n8n community node for the Nexara speech-to-text API: transcription, diarization, speaker roles, emotion recognition, and structured LLM output

An Eve extension that exposes a ChatGPT Enterprise user's authorized connectors as dynamically discovered tools.

A Codex-style checkpoint compaction strategy for Eve agents.

Arabic-First PDF Extraction, RTL Normalization & Table Chunking Engine for RAG Systems.

n8n community node for Claix — extract PDF, Word and Excel documents to structured JSON

Regression testing and configuration sweeps for RAG pipelines.