RunanywhereAI/RCLI

Health v2 · maintenance only

Adoption not part of the score

1542 stars · 84 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

RCLI is a single-binary CLI that runs open-source AI models locally on your machine, covering chat, vision, speech-to-text, text-to-speech, and embeddings with no cloud dependency. It dispatches to multiple inference backends (llama.cpp, MLX, Sherpa-ONNX, ONNX Runtime, and private Core ML/QNN overlays) and can serve an OpenAI-compatible API.

Use cases

• run local llm chat on my mac without cloud
• query my documents with on-device rag
• transcribe speech to text offline
• generate speech from text locally
• serve an openai-compatible local model api
• pull and run gguf models from hugging face
• generate images locally on apple silicon

When to choose

• you want fully private, offline AI inference on Apple Silicon or desktop hardware
• you need one binary that handles LLM, speech, embeddings, and vision across multiple backends
• you want an OpenAI-compatible local endpoint for existing tooling

When to avoid

• you need large-scale server or multi-GPU cluster inference
• you rely on proprietary frontier models not available as open weights
• you need Windows ARM64 or Linux GPU features that depend on private overlay engines

Facets

cli-tool · maturity active

llm-inference speech-recognition tts rag chatbot http-server cli artificial-intelligence large-language-models speech-processing developer-tools privacy windows cli on-device-ai local-llm apple-silicon metal llama-cpp mlx gguf voice-assistant openai-compatible-api offline command-line macos linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="RunanywhereAI/RCLI")