AXIOMINFERENCE Get Axiom

AxiomInference

The AI Lab for
local‑first
intelligence.

AxiomInference builds practical AI tools that run entirely on your own hardware, and researches small model architectures that adapt instead of just scaling. No subscriptions. No data leaving your device.

Latest release

Axiom V1.8.6

A Windows local-first AI assistant and agentic workspace with local GGUF, self-hosted Hybrid Local, optional OpenRouter models, persistent chat, tools, Skills, Plugins, and Workplace modes.

Continue reading

Latest releases

All repositories →
AXIOM · V1.8.6 Local-first workspace with Hybrid Local and Workplace Use local GGUF, a self-hosted OpenAI-compatible endpoint, or optional cloud models with persistent tools, Skills, and agent workflows. AXIOM CLI · v1.15.0 Axiom Code-powered terminal agent Use the Axiom council or an OpenCode-powered runtime with local tools, shell access, tests, Git, Kestrel 1, and automatic context compaction. RESEARCH · NANO V0.1 TRAINED Project Kestrel — trained small-model research Kestrel-Nano v0.1 is a trained 157M-parameter model; Phase C's 4.86B-token dataset and evaluation work are complete, with Phase D next.

What we build

Two products, one active research program.

Axiom and Axiom CLI are local-first AI tools: a Windows desktop assistant and a cross-platform terminal coding agent. Both support your own self-hosted OpenAI-compatible endpoint alongside local and optional cloud inference.

Project Kestrel is the research program behind a from-scratch, small-model architecture. Kestrel-Nano v0.1 has been trained; current work is focused on stronger data, evaluation, and the next over-training phase.

Why local-first

You own the hardware. You own the data.

Most AI tools send your conversations to a server. Axiom runs local GGUF models on your own machine by default, with optional cloud models when you want more power — always with your own API key, never a subscription.

Every product here also supports pointing at your own self-hosted inference server, so the same philosophy extends to hardware you run yourself, anywhere.