Documentation

Set up SharpMind anywhere.

The same Studio, the same model format, four places to run it. Start with the machine in front of you; add a GPU host when a fine-tune needs more memory. Every guide includes the errors you are likely to see and what they mean.

Download for Windows Set up a GPU host → Start locally

GPU · SSH

Vultr + SSH setup

Create the instance, add one key, and let Studio install Docker, the NVIDIA toolkit and the server image.

recommended for a first GPU run Local

Local Docker

Build the Compute image on this PC (or the GPU server image) and start it from Settings.

no account · CPU or local NVIDIA GPU Cloud

Runpod

Deploy a GPU pod through the Runpod API and terminate it when the run is done.

API key · Docker Hub · proxy URLs Models

HuggingFace

Browse GGUF repositories, check licence and size, download with resume and convert to .smm.

public downloads · token for gated Remote models

OpenRouter

Connect hosted models for Chat and Work with an encrypted key and local tool permissions.

inference only · per token Projects

Work mode

Project context, scoped file and web tools, skills and explicit read/write/delete permissions.

local or OpenRouter Access

Phone pairing

Open the local Studio from a phone on your LAN with a QR or PIN, then revoke the device when you are done.

LAN only · revocable tokens Overview

Compare providers

Local Docker, Runpod, Vultr or your own Linux box: cost, ports and who runs the setup.

start here if you are unsure Train

Pretrain, LoRA and SFT

Which jobs use the GPU, which stay on the CPU, and why a LayerNorm base falls back.

GPU fine-tuning: LoRA only Errors

Troubleshooting

Pull access denied, accelerator declined, firewall timeouts, SSH auth, 401 after a token rotation.

the exact messages

What works today

The support matrix.

This PC covers the Studio process and an optional Local Docker host on the same machine. A local NVIDIA GPU is used only when the GPU image and Docker runtime are present. A Compute host always brings its own device.

Capability	This PC / Local Docker	Compute host	Notes
Chat with a packaged model	Yes	Yes	One active model per process on a host.
OpenAI-style API	—	Yes	Bearer token; /v1/chat/completions, models, files, training jobs.
Pretrain from scratch	Yes (CPU)	Yes (CPU)	The GPU engine refuses whole-weight training by design.
LoRA fine-tune	Yes (GPU or CPU)	Yes (GPU or CPU)	GPU needs RMSNorm + RoPE/NoPE + gated FFN; otherwise the CPU retry runs.
SFT (assistant-only loss)	Yes (GPU or CPU)	Yes (GPU or CPU)	Chat JSONL; the base supplies the tokenizer and template.
GPU inference	With a supported local GPU	Yes (GPU or CPU)	Also covers LayerNorm and dense-GELU models.
Remote data and artifacts	—	Yes	Uploads with progress; exports download back to Studio.

Version 1.1.0. The repository is the source of truth. The server image contract is stable; later releases may still adjust screens or flags.

FIVE-MINUTE PATH

Just want to see it run?

Build the app, pick a GGUF or .smm, and chat locally. No account and no host.

Build from source

GPU PATH

Want a real fine-tune?

Rent an A16, add one SSH key, and let the Compute card install and deploy the GPU image.

Vultr guide →