One API key, one endpoint — access to modern AI models
Thelsy is an API gateway service. We provide software developers, indie builders and small teams with programmatic access to large language models through a single, OpenAI-compatible endpoint. Point your existing SDK at our base URL and start building — no infrastructure to run, no provider accounts to manage, no token-by-token bookkeeping to write yourself.
What the service is
OpenAI-compatible API
Standard /v1/chat/completions interface. Works with official OpenAI SDKs and
common third-party clients by changing only the base URL.
Token-based usage, monthly plans
Subscribe to a monthly plan that includes a token allowance. Usage beyond the allowance is billed per million tokens at the overage rate shown below.
Usage dashboard & API keys
Create and revoke your own API keys, set per-key limits, and monitor consumption and spend in your account dashboard.
Vision input supported
The model we serve accepts text and image input (image understanding). It does not generate images.
Base URL
https://api.thelsy.com/v1
Example request:
curl https://api.thelsy.com/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"deepseek-flash","messages":[{"role":"user","content":"Hello"}]}'
Pricing
All plans are billed monthly in US dollars. Included allowance is measured in tokens (one million tokens = 1M tokens). Cancel any time; plans renew until cancelled.
| Plan | Monthly price | Included usage | Overage |
|---|---|---|---|
| Starter | $19 / month | 30M tokens | $0.80 / 1M tokens |
| Pro | $49 / month | 80M tokens | $0.80 / 1M tokens |
| Business | $149 / month | 250M tokens | $0.75 / 1M tokens |
How it works
- Choose a plan and complete checkout on our payment platform.
- You receive access to your account, where you create one or more API keys.
- Call
https://api.thelsy.com/v1from your application with your key. - Track token usage and spend in your dashboard.
Who it is for
- Developers building chat, summarisation, extraction or analysis features into their products
- Small SaaS teams that want one API key instead of multiple provider accounts
- Agencies and freelancers who need reliable throughput for client projects
Capability boundaries
- Supported: text generation and conversation; image understanding (image input).
- Not supported: image generation; audio/video generation; results are produced by the upstream model provider and may be inaccurate — verify important output.
Acceptable use
This is an API access service for lawful software development. You may not use it to generate illegal content, to deceive or defraud, to send spam, to infringe intellectual property, to violate privacy, or to attempt to bypass model safety systems. A full list of prohibited uses and the terms governing the service are in our Terms of Service.
Contact
Questions before subscribing, or need help after? Email 393516700@qq.com. We usually reply within one business day.