Chat & reasoning
DeepSeek V4 Pro, Qwen 3.8, Kimi K2.6, GLM-5.3 and fast Flash models for coding, agents, writing and translation.
For indie developers, makers and freelancers
Top up once.
Use every model.
Chat, write, code and create images and video with DeepSeek, Qwen, Kimi, GLM, Seedance and more in one workspace. Pick the models you want, see the cost before you send, and pay from one balance.
- Plans from $30
- No company paperwork
- Cancel anytime
All lines, one key
- DeepSeekV4 Pro · V4.1 Flash
- QwenQwen 3.8 · Embed
- GLMGLM-5.3 · Flash
- KimiKimi K2.6
- FLUXText to image
- SeedanceText to video
How it works
From sign-up to first request in three stops
- 1
Sign up with email
No company name, tax ID or approval queue. Confirm your email and your account is ready.
- 2
Pick a plan
Choose Plus, Pro or Ultra and pay by card. Your plan fee becomes usage credit, and you can top up anytime.
- 3
Pick models and go
Use one model or compare up to three side by side. Every message shows its cost before you send it.
Models
Chat, code, images and video
Popular open models from leading labs, plus image, video and speech models, all paid from one balance.
Images
FLUX.1 schnell for quick drafts and FLUX.1 dev for higher-quality images.
Video
Seedance text-to-video: five-second 720p clips, with the maximum price fixed before generation starts.
Compare side by side
Send one prompt to up to three models and keep the answer you like best, with each answer's cost next to it.
Up to 3 models per message
Cost before you send
The estimate updates as you type and includes the chat history, so long chats never surprise you. The exact cost appears under each answer.
Estimate first, actual cost after
Not sure which to pick?
AI Master Fast and AI Master Pro always point to our recommended models, and we upgrade them when a better one arrives.
AI Master Fast · AI Master Pro
Pricing
Your plan fee comes back as credit
Every dollar you pay for a plan becomes usage credit. Bigger plans add higher limits and bonus credit on top-ups. Change or cancel anytime.
Top up anytime
Add credit by card whenever you need it. Top-up credit stays valid for 12 months and is used after your plan credit.
Prices in US dollars. Local taxes may be added at checkout. Plan credit renews each billing month and does not roll over.
Pay-per-use prices
Models & prices
Per 1M tokens unless noted. Usage beyond your plan credit is charged at these prices.
| Model | Type | Where it runs | Context | Input | Output |
|---|
The workspace
Pick models. See the cost. Send.
Everything happens in your browser. No code, no API keys, no surprises on the bill.
- Compare up to three models side by side
- Estimated cost before you send, actual cost after
- Images and 5-second video clips in the same place
- Chats stay in your browser; we keep only usage records for billing
Your data
You always see where your requests run
Every model shows its data zone before you pick it, and every reply shows where it ran. Most models today run in the External zone: the model makers' own services, including in China.
regionalMore models soon
Regional zone
Suppliers inside our Regional zone, with the strongest data control. We're adding models here; they're labelled regional when available.
externalMost models today
External zone
Suppliers outside it, such as the model makers' own services, including in China. The widest choice and the lowest prices. Don't send personal or confidential data to these models.
metadataMetadata only
Prompts aren't stored
We record usage metadata for billing: time, model, tokens, cost and latency. We don't store your prompts or the model's answers.
Each response carries x-data-zone and x-processing-region headers, and your usage log shows where every request ran.
Need company invoices, monthly billing or a private deployment?
AI Master for BusinessWhy AI Master
Built so a side project never surprises you
Automatic failover
Popular models have more than one supplier. If one fails, your request moves to the next one in your allowed zones, and you're charged once.
No overspending
We hold a request's maximum cost, then charge the real cost and release the rest. Parallel requests can't overdraw your balance.
Spend caps per key
Give each app its own key with a monthly cap and a list of allowed models. Revoke a key at any time.
Usage down to the request
Tokens, cost and latency for every call, filterable by key and model, with CSV export.
Failed requests are free
You pay only for requests that succeed, at the published price.
A receipt for every payment
Each top-up and plan payment gets a numbered receipt you can print or save as PDF.
FAQ
Questions, answered
Do I need a company to sign up?
How do plans and top-ups work together?
Does unused credit roll over?
Can I change or cancel my plan?
Which payment methods do you accept?
Do I need to code or set up API keys?
Where is my data processed? Do you store prompts?
What happens if a model provider goes down?
Can I get a refund?
I need invoices or monthly billing for my company.
Your workspace is a minute away
Plans from $30 a month. Top up anytime and switch plans whenever you like.