# Start here **URL:** https://forum.hal0.dev/c/docs/docs-getting-started/19.md [Latest](https://forum.hal0.dev/latest.md) · [Categories](https://forum.hal0.dev/categories.md) · [Tags](https://forum.hal0.dev/tags.md) --- ## [Hal0 docs: Start here](https://forum.hal0.dev/t/hal0-docs-start-here/59)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:51am UTC](https://forum.hal0.dev/t/hal0-docs-start-here/59 "2026-08-22T06:51:15Z")
Start here hal0 documentation: Hal0 documentation Install hal0: Install hal0 (hal0 docs) GPU drivers and memory: GPU drivers and memory Install on bare metal: Install on bare metal Install on Proxmox: Install on Proxmox … --- ## [About the Start here category](https://forum.hal0.dev/t/about-the-start-here-category/80)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 28, 2026, 8:31pm UTC](https://forum.hal0.dev/t/about-the-start-here-category/80 "2026-08-28T20:31:28Z")
Install, first model, first chat. Bare metal, Proxmox, WSL. --- ## [Install hal0 (hal0 docs)](https://forum.hal0.dev/t/install-hal0-hal0-docs/20)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/install-hal0-hal0-docs/20 "2026-08-22T06:49:17Z")
hal0 installs from a single command. The bootstrap script is the trust boundary: it fetches the release manifest, verifies the tarball, unpacks it, and hands off to the real installer (installer/install.sh). curl -fsS… --- ## [Install on Proxmox](https://forum.hal0.dev/t/install-on-proxmox/21)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/install-on-proxmox/21 "2026-08-22T06:49:19Z")
Proxmox VE is a common host for hal0, especially on AMD Strix Halo (Ryzen AI Max+) hardware where an LXC shares the host kernel and the iGPU + XDNA NPU pass through as ordinary device nodes — no VM, no vGPU shim, near-… --- ## [Install in an Incus container](https://forum.hal0.dev/t/install-in-an-incus-container/132)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [November 1, 2026, 8:40pm UTC](https://forum.hal0.dev/t/install-in-an-incus-container/132 "2026-11-01T20:40:22Z")
A bare-metal install puts hal0 deep into the host: systemd units, a service user with sudoers grants, a rootful podman store, Avahi services and a self-updater that owns /usr/lib/hal0. On a dedicated appliance that is… --- ## [Hal0 documentation](https://forum.hal0.dev/t/hal0-documentation/19)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/hal0-documentation/19 "2026-08-22T06:49:15Z")
hal0 turns a Linux box into a polished, OpenAI-compatible inference appliance. It is built for the AMD Ryzen AI Max+ 395 (“Strix Halo”) — iGPU + XDNA NPU + 128 GB of unified memory — and falls back to portable hardwar… --- ## [Install on bare metal](https://forum.hal0.dev/t/install-on-bare-metal/15)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/install-on-bare-metal/15 "2026-08-22T06:49:07Z")
Bare metal is the simplest way to run hal0: the installer handles the container runtime and Python venv for you, and on a Strix Halo box the iGPU and NPU are already present as local device nodes — no passthrough to c… --- ## [Load your first model](https://forum.hal0.dev/t/load-your-first-model/18)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/load-your-first-model/18 "2026-08-22T06:49:13Z")
The installer seeds slots and starts services — there is no setup wizard in v1.0. Models in hal0 live in a local registry, the single source of truth for Hugging Face coordinates, SHA-256 digests, and curated filenames… --- ## [Send your first chat](https://forum.hal0.dev/t/send-your-first-chat/17)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/send-your-first-chat/17 "2026-08-22T06:49:11Z")
With a model pulled and assigned to a slot, you can chat through the prewired web UI or directly against the OpenAI-compatible API. With OpenWebUI hal0 installs and prewires OpenWebUI, which binds port 3001 and points… --- ## [Verify the install](https://forum.hal0.dev/t/verify-the-install/109)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [October 2, 2026, 6:16pm UTC](https://forum.hal0.dev/t/verify-the-install/109 "2026-10-02T18:16:22Z")
install.sh finishes by starting hal0-api and printing a summary. Before you trust the box, run these six checks in order — each one rules out a different layer (network, model registry, inference, UI, logs, and the ag… --- ## [Quick start — WSL2 (experimental)](https://forum.hal0.dev/t/quick-start-wsl2-experimental/22)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/quick-start-wsl2-experimental/22 "2026-08-22T06:49:21Z")
▶ Danger: Experimental and unsupported What to expect WSL2 runs a real Linux kernel, so the hal0 control plane, podman, and CPU inference can work — but the accelerator story does not translate: Capability Under WSL… --- ## [GPU drivers and memory](https://forum.hal0.dev/t/gpu-drivers-and-memory/16)
**Author:** [@system](https://forum.hal0.dev/u/system)\ **Replies:** 0\ **Last updated:** [August 22, 2026, 6:49am UTC](https://forum.hal0.dev/t/gpu-drivers-and-memory/16 "2026-08-22T06:49:09Z")
hal0’s inference containers need the host kernel to already expose the GPU (and, on Strix Halo, the NPU) as device nodes. hal0 ships the userspace runtime — ROCm, XRT, the FastFlowLM binary — inside its container imag…