What is NeuralCleave?
NeuralCleave is an open-source personal AI assistant that brings every AI model, conversation, memory, and workflow together in one private, extensible intelligence layer. Rather than tying you to a single app, it runs one assistant across 32 channels — Telegram, WhatsApp, Discord, Slack and 29 more — and 13 LLM providers including Anthropic, OpenAI, DeepSeek and Ollama, so you reach the same assistant wherever you already chat. Its 3-tier memory system uses Redis for hot session data, Qdrant for vector semantic recall, and SQLite for long-term storage, with per-agent namespace isolation, compaction and an archiver, and it works fully offline. Task-aware model routing automatically sends 10 task types to the optimal provider — complex reasoning to Claude Opus, code to DeepSeek, cheap inference to Ollama — and a privacy mode can force all calls to run locally. NeuralCleave is voice-ready with Whisper speech-to-text, 3-tier text-to-speech and wake-word support across platforms, and it ships a plugin SDK with typed abstract base classes, entry-points, hot-reload and a Hub marketplace. The current v2.1.5 release advertises 41 REST endpoints and thousands of passing tests. It fits developers and privacy-minded power users who want a self-hosted, model-agnostic assistant they fully control rather than a closed cloud chatbot.
What is NeuralCleave used for?
Best for developers and privacy-minded power users who want a self-hosted, model-agnostic AI assistant across many chat channels.
How much does NeuralCleave cost?
Open source
NeuralCleave is open-source and self-hosted; you supply your own infrastructure and pay the API costs of the LLM providers you connect
Hosted / paid
No hosted or paid plan pricing is published on the homepage
Pros and cons
Pros
- One assistant across 32 channels (Telegram, WhatsApp, Discord, Slack and more) and 13 LLM providers
- 3-tier memory: Redis hot session, Qdrant vector semantic, SQLite long-term, with per-agent isolation and offline support
- Task-aware routing sends 10 task types to the optimal model (reasoning to Claude Opus, code to DeepSeek, cheap inference to Ollama)
- Voice-ready with Whisper STT, 3-tier TTS and wake word; privacy mode forces all calls to run locally
- Open-source with a plugin SDK (typed ABCs, hot-reload, Hub marketplace) and 41 REST endpoints
Cons
- Self-hosting requires running Redis, Qdrant and SQLite, so setup is heavier than a hosted chatbot
- No pricing is published; value depends on your own infrastructure and the API costs of the providers you route to
- Broad multi-channel, multi-provider scope means a steeper learning curve for non-technical users
- At v2.1.5 it is still a relatively young project despite a large test suite
Questions about NeuralCleave
What is NeuralCleave?
NeuralCleave is an open-source personal AI assistant that unifies models, conversations, memory and workflows in one private layer. It runs across 32 channels and 13 LLM providers, uses a 3-tier memory system, routes tasks to the best model, and supports voice and a plugin SDK.
Is NeuralCleave free and open source?
Yes. NeuralCleave is open-source and self-hosted, so the software itself is free to run. You provide your own infrastructure (Redis, Qdrant, SQLite) and cover the API costs of any cloud LLM providers you route to; local models via Ollama can avoid those API costs.
Which channels and LLM providers does NeuralCleave support?
NeuralCleave runs one assistant across 32 channels, including Telegram, WhatsApp, Discord and Slack, and connects 13 LLM providers such as Anthropic, OpenAI, DeepSeek and Ollama. It routes 10 task types to whichever provider fits each task best.
How does NeuralCleave handle memory?
NeuralCleave uses a 3-tier memory system: Redis for hot session data, Qdrant for vector semantic recall, and SQLite for long-term storage. It adds per-agent namespace isolation, compaction and an archiver, and can operate fully offline.
Is NeuralCleave private?
It is built for privacy: it is self-hosted, and a privacy mode forces all model calls to run locally via Ollama so data never leaves your machine. Memory is stored in your own Redis, Qdrant and SQLite instances rather than a third-party cloud.
What are the best alternatives to NeuralCleave?
Reviews
Sign in to rate and review NeuralCleave.
Sign in to reviewNo reviews yet. Be the first to review NeuralCleave!