AIHostBox
Client area Deploy an app
All 57 applications
CategoriesAI assistants & chatLLM runtimesAgents & RAGGateways & LLM opsVector & dataAutomationSpeech, image & videoSearch & web agentsSupport & CXML platformMCP servers
SolutionsPrivate team chatUnlimited automationCompany knowledge AIAI gateway & observability
More PricingGuides ComparisonsToolsPrivate LLM hosting LicensingSecurity SupportClient area

Text Generation WebUI hosting

Oobabooga’s all-in-one interface for running and testing local models.

LLM runtimesAGPL-3.0from $16.99/mo
In short

Text Generation WebUI hosting at AIHostBox means a dedicated, single-tenant Text Generation WebUI instance on our infrastructure: we install it, pin it to a reviewed release, put it behind TLS on your domain, back it up nightly and keep it patched. Plans start at $16.99 per month and the admin account, the data and any API keys are yours.

Multiple loadersNotebook modeLoRAExtensionsAPI

Text Generation WebUI — oobabooga’s project, licensed AGPL-3.0 — is a Gradio interface for loading and probing language models. It supports several backends (llama.cpp, ik_llama.cpp, Transformers, ExLlamaV3 and TensorRT-LLM) and switches between them without a restart. Chat, chat-instruct and instruct modes sit beside a notebook tab for raw completion, and every sampler parameter is on screen. Where Ollama hides the knobs to make serving simple, this project exposes all of them, which is why people use it to compare models and settings before committing to one.

Running it on a server rather than a laptop changes two things. The machine holding the models is always on, so the OpenAI- and Anthropic-compatible API on port 5000 can back other tools — a chat front-end, a script, an agent — while you keep the Gradio UI for experiments. And the model folder lives in one place: download a GGUF once into user_data/models and every device you use reaches the same weights, presets and saved chats, rather than each machine keeping its own copy.

Text Generation WebUI hosting plans

StarterRuns 7-8B models (Q4)
8 GB RAM · 4 vCPU · NVMe
$16.99/mo
Order
PlusRuns 13B models (Q4)
12 GB RAM · 6 vCPU · NVMe
$27.99/mo
Order

Every plan is a dedicated single-tenant instance with NVMe storage, nightly off-box backups, TLS and managed patching. Quarterly, semi-annual and annual terms are discounted at checkout. Model tokens are not included — you bring your own key.

How we run Text Generation WebUI

A single Text Generation WebUI container built from the project’s CPU Docker configuration, with the Gradio UI on port 7860, the API on 5000 and user_data on persistent storage.

What we configure for you

  • --listen with --gradio-auth so the UI is reachable only over TLS with your credentials
  • --api and --api-key for the OpenAI-compatible endpoint on port 5000
  • llama.cpp thread count matched to the plan’s vCPUs
  • user_data/models reachable over SFTP for uploading GGUF files
  • Built-in extensions enabled on request

Which plan fits

Starter (8 GB, 4 vCPU) runs 7-8B models at Q4 at roughly 5-15 tokens per second, enough for evaluation and light API use. Plus (12 GB, 6 vCPU) fits 13B models at Q4 with room for context; anything larger is not realistic on CPU.

The honest limits

This is CPU hosting, so the GPU-only loaders (ExLlamaV3, TensorRT-LLM) are unavailable and long contexts slow generation noticeably. The Gradio login is shared, with no per-user accounts.

What people run Text Generation WebUI for

Model evaluation

Load several models and compare behaviour with identical settings.

Sampler tuning

Find the generation settings that suit your use case before deploying.

Private API backend

Serve a GGUF model on an OpenAI-compatible endpoint for your own scripts and front-ends.

Drafting in notebook mode

Iterate on raw completions without chat templating getting in the way.

Licence and version policy

ItemPosition
Upstream licenceAGPL-3.0
Sold as managed hostingYes
TenancySingle tenant — one customer per instance
PatchingCritical advisories within 24 hours, high severity within 72

Our full position on licences, including the applications we decline to sell, is on the licensing page.

Works well with Text Generation WebUI

Open WebUI logo

Open WebUI

A ChatGPT-style workspace for your team, pointed at any model you like.

from $5.99/moDetails →SillyTavern logo

SillyTavern

The character-driven chat front-end, online wherever you are.

from $5.99/moDetails →LiteLLM logo

LiteLLM

One OpenAI-compatible endpoint in front of every model provider you use.

from $5.99/moDetails →

More LLM runtimes we host

Ollama logo

Ollama

Your private LLM server — pull a model, get an API, keep every token on your own box.

from $16.99/moDetails →Hermes logo

Hermes

Nous Research’s Hermes models, served privately behind your own API.

from $16.99/moDetails →LocalAI logo

LocalAI

A drop-in OpenAI replacement for text, images, audio and embeddings.

from $16.99/moDetails →KoboldCpp logo

KoboldCpp

A single-binary GGUF server with its own UI and a KoboldAI API.

from $13.99/moDetails →
Questions

Text Generation WebUI hosting: your questions

Which model formats run on a CPU plan?
GGUF files through the llama.cpp or ik_llama.cpp loaders. ExLlamaV3 and TensorRT-LLM need an NVIDIA GPU, and full-precision Transformers models are too large and too slow on CPU to be useful. In practice that means a Q4 quant of a 7-8B model on Starter and a 13B model on Plus.
How do I use the API?
The --api flag opens chat, completions and an Anthropic-style Messages endpoint on port 5000, with tool calling. We enable it on request, protect it with --api-key and serve it over TLS, so any OpenAI SDK works once you change the base URL and add your key.
How is the web interface protected?
Text Generation WebUI has no user accounts. Access is a Gradio login set with --gradio-auth as username:password, which we configure and serve only over HTTPS. Everyone with those credentials shares the same models, presets and chat history, so treat it as a single-person or single-team workbench.
How do I add my own models?
Upload GGUF files over SFTP into user_data/models, or paste a Hugging Face repository name into the download field on the Model tab. The model and its context window both have to fit in the plan’s RAM, so pick quantisation to suit Starter’s 8 GB or Plus’s 12 GB.
What does the backup cover?
The user_data directory: models, saved chats, characters, presets and settings. Model files make up most of its size, but they can be downloaded again; the chats, characters and tuned presets are the part you cannot recreate, and they are included in every nightly off-box backup.

Run Text Generation WebUI without running a server

A dedicated Text Generation WebUI instance, deployed and maintained by us. Month to month, cancel whenever.

Browse the catalogue Talk to a human