Ollama runs language models locally and provides tools to work with model files and APIs.
Open source
24 alternativesRunAnywhere
The RunAnywhere SDK repository describes a production toolkit for running AI locally.
About RunAnywhere
RunAnywhere is a GitHub repository for a toolkit that runs AI models locally on phones, browsers, desktops and servers. It describes eight SDKs built on one C++ core, with a capability registry that routes each call to the engine best suited to the device, plus a Console for managing a fleet. Documented capabilities include LLM chat, structured JSON output, tool calling, vision-language understanding, speech-to-text, text-to-speech, voice agent pipelines, embeddings, local retrieval-augmented generation and image generation.
Key features
- On-device LLM chat with token streaming, multi-turn history and LoRA adapters
- Speech-to-text via Whisper and Moonshine; TTS via Piper, Kokoro, Kitten and others
- Schema-validated JSON output and local tool calling with an agent loop
- Vision-language image understanding and local document RAG with streaming
- Engine routing across Hexagon NPU, Apple MLX, llama.cpp, ONNX and Core ML
- Python package and command-line interface for running models from a terminal
Best for: Developers building mobile, web, desktop or server applications that need AI inference to run on the device rather than in a remote service.
Good for
- Described as private by default and offline by design, with models downloaded on first use.
- One semantic API is exposed across eight SDKs and multiple hardware-specific engines.
- A capability query reports what the current package and device can actually execute.
Watch out for
- Wake-word detection is not implemented, and the computer-use action parser is not a full autonomous agent framework.
- Some capabilities are platform- or backend-gated, such as image inpainting on the Hexagon NPU, and LiteRT and ExecuTorch are reserved values rather than integrated runtimes.
- Enum presence for a capability does not mean the matching engine is installed.
RunAnywhere at a glance
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
RunAnywhere alternatives
Tools and sites you could use instead, ranked by how closely they match.
This repository contains utilities intended for use with Llama models.
DeepEval is an LLM evaluation framework, according to its repository description.
Open source
24 alternatives
Stakpak runs an agent on a machine to ship code and keep applications running.
Open source
24 alternatives
Glide provides an open model gateway for developing production generative AI applications.
Open source
24 alternativesVercel’s AI SDK is a TypeScript toolkit for building AI applications and agents with React, Next.js, Vue, Svelte, and Node.js.
Popular in AI Infrastructure & MLOps
Well-known listings from the same topic.
Guild AIguild.aiFlow AIflow-ai.comsystemprompt.iosystemprompt.ioCoreWeavecoreweave.comKili Technologykili-technology.comAPI docsdocs.x.ai
RunAnywhere FAQ
Does inference run fully on the device?
The toolkit is positioned for local, offline execution; the Python quick-start example notes that a model is downloaded on first use.
Does the voice pipeline include wake-word detection?
No. The voice agent pipeline covers VAD, STT, LLM and TTS, but wake-word detection is stated as not implemented.
How can I tell what a given device can run?
Call RunAnywhere.capabilities() (v4) to discover what the current package and device can execute; the presence of an enum value alone does not mean an engine is installed.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.