from 12 GBChat
Private ChatGPT · MiMo 9B
A fast chat from Xiaomi on an affordable card. It understands images too: send a photo or a screenshot.
from €0.12/h · RTX 3060, 12 GB
Ready-made app · Qwen3
Run Qwen3 on a rented GPU: your private ChatGPT
The ready-made Private ChatGPT app installs Ollama, a Qwen3 model and the Open WebUI chat window. Qwen3 32B runs on a 24 GB card and writes well in many languages; Qwen3 8B runs on 12 GB and answers faster. Your conversations stay on your server.
- App
- Private ChatGPT · Qwen3 32B
- GPU
- RTX 3090, 24 GB
- Price
- from €0.30/h
To launch, your balance must cover one hour of the machine. You are billed by the minute for actual use. The server installs Private ChatGPT · Qwen3 32B for you.
from 24 GBof video memory
2–5 minutesuntil the server is up
By the minutebilled from your balance
Open WebUIopens in your browser
What is on the server after launch
There are two versions. Both install Ollama and the chat window for you: Qwen3 32B is more accurate, Qwen3 8B runs on a cheaper card.
Private ChatGPT · Qwen3 32B
from 24 GBA chat in your browser for writing, code and documents. No subscriptions and no message limits.
- Open WebUI: a chat window in your browser
- Ollama with the Qwen3 32B model
- The model answers on the server itself, no external API
from €0.30/h · RTX 3090 Launch Qwen3 32B
Private ChatGPT · Qwen3 8B
from 12 GBA fast chat on an affordable card: email drafts, translations, quick answers.
- Open WebUI: a chat window in your browser
- Ollama with the Qwen3 8B model
from €0.12/h · RTX 3060 Launch Qwen3 8B
Licence Apache 2.0: you can use the answers and code in your work without restrictions.
How much memory you need and which GPU to pick
Private ChatGPT · Qwen3 32B needs a card with at least 24 GB; the cheapest suitable one is the RTX 3090. The table shows memory for other versions and tasks. Prices update from available catalogue machines when you open the page.
| Version | FP16 | 4-bit | Suitable GPUGPU | Price per hourPer hour |
|---|---|---|---|---|
| Qwen3 8B | 18 GB | 6 GB | RTX 3060 | €0.12 |
| Qwen3 14B | 31 GB | 10 GB | RTX 3060 | €0.12 |
| Qwen3 32B | 71 GB | 22 GB | 2× RTX 3060 | €0.24 |
Estimate: FP16 ≈ 2 GB per billion parameters, 4-bit ≈ 0.6 GB, plus 10% for context. Pick: the cheapest configuration with enough memory; for large models that means several identical cards. All GPUs and prices
How to run a private ChatGPT on Qwen3 in 3 steps
- Sign up and top up. Create an account and top up from €10. To launch, your balance must cover at least one hour of the chosen machine; after that you pay by the minute.
- Launch the app. Click Launch Qwen3 32B. The dashboard opens with Private ChatGPT · Qwen3 32B and a card with at least 24 GB already selected; check the price and confirm.
- Work in your browser. The server starts in 2–5 minutes, and the app downloads the model in a few more. Then press the Open WebUI button in the dashboard and type your prompt.
Finished? Stop the server in the dashboard and billing stops. Save and stop keeps /workspace for €0.03 per GB per month, and it comes back on your next server automatically.
Commands for a manual install on an empty server
Ollama and Open WebUI
curl -fsSL https://ollama.com/install.sh | sh
ollama pull qwen3:32b
pip install open-webui
open-webui serve --port 3000
What people run a private ChatGPT on Qwen3 for
01
Writing and email
Email drafts, posts, product descriptions, translation.
02
Code
Explain someone else’s code, find a bug, write a function or an SQL query.
03
Documents
Summarise a contract or a report, answer questions about a file.
Tips before you launch
- Qwen3 reasons first and answers second. For a quick answer, add /no_think to your message.
- Drag a PDF or a text file into the chat window and ask about it.
- The app uses an 8k token context, so 32B fits entirely in 24 GB and answers quickly.
Other ready-made apps
Each app installs its interface and model on its own. You only pay for GPU time.
from 12 GBText to speech
Text to speech · Qwen3-TTS
Turns text into speech in English and other languages, with several ready-made voices.
from €0.12/h · RTX 3060, 12 GB
Frequently asked questions
How is Qwen3 32B different from 8B?
32B is more accurate on hard questions, code and long texts and needs a 24 GB card. 8B runs on 12 GB and is fine for emails, translations and short answers.
Where do my messages go?
Nowhere: the model runs on your rented server, with no external API. If you stop the server without saving, the disk and your conversations are deleted.
How long does a launch take?
The server usually starts in 2–5 minutes. Then the app downloads the model on its own, which takes a few more minutes, and a button that opens Open WebUI appears in the dashboard.
What happens to my models and results after I stop?
When you stop, click Save and stop: the /workspace folder with your models, LoRAs and results goes to storage for €0.03 per GB per month and comes back on your next server automatically. If you stop without saving, the disk is deleted.
How is it billed?
By the minute from your balance in euros. Top up from €10; to launch, your balance must cover at least one hour of the chosen machine. Stop the server and billing stops.
A server is usually ready in 2–5 minutes
Sign up in a minute, top up from €10, pay by the minute. Stop the server and billing stops.