Build more.
Spend less.

DeepSeek, Qwen, GLM, Kimi, MiniMax and MiMo, so you can pick the right model for the job and the budget.

DeepSeekQwenGLMKimiMiniMaxMiMo
The TexAPI dashboard: your credit balance, usage and per-token rates
Checking statusFour industry-standard API interfaces
Pricing

Pay for what you use. See where every credit goes.

Top up to receive credits and spend them per request, or take a monthly plan that adds credits to the same balance and discounts usage on top. Usage value is shown in credits; actual payments remain in USD.

Loading pricing…

—

Pick the model for the job — and for your budget.

Every card shows its own input and output rates. Rates below are what you actually pay.

Checking status
Loading catalogue…
Rates are per 1M tokens, before any subscription discount.Full catalogue
Tool compatibility

Eight coding tools, connected out of the box — one catalogue, four API interfaces.

Claude Code, Codex CLI, Gemini CLI, Cline, Roo Code, Continue, aider and Zed all work with the families on sale; Kimi serves Messages natively, the rest serve every interface. The full matrix lives on the Docs page.

See the tool × model matrix

—

Know what your month costs before you spend it.

Pick a model and a monthly token volume. We calculate it at the credit rate on that model's card and pick the better option — pay as you go or a plan — so you know the credits your workload will use.

Loading rates…

FAQ8 questions

Frequently
Asked Questions

How billing, limits and compatibility actually work.

  • DeepSeek, Qwen, GLM, Kimi, MiniMax and MiMo. The Models section above lists every model we serve together with its live per-token price, so you can pick the right model for the job and the budget. No model is locked behind a higher tier — every plan reaches the whole catalogue.

  • Each request uses credits according to its token count and the model's displayed credit rate. Top up to receive credits or take a monthly plan that adds credits and discounts usage. 100 credits represent $1 of usage value.

  • No. Credits stay in your balance until you use them, including credits added by a monthly plan. Nothing resets at the end of a period.

  • Requests per minute, tokens per minute and the number of API keys you can hold all depend on your plan. The Subscription tab under Pricing lists the exact limits for each tier.

  • Yes. Point your OpenAI SDK's base_url at our endpoint. We also serve the Anthropic Messages and Gemini formats on the same key, so Claude Code, Cursor and Cline work after changing one environment variable — no new setup to build first.

  • On models that support it, cached input is billed at roughly a tenth of the normal input rate. Each model in the Models section shows its cached rate where one exists.

  • Card and PayPal through Creem, or bank transfer through payOS. Both add credits to the same balance at 100 credits per USD paid.

  • Prices exclude VAT. The system does not add VAT at checkout today; where VAT applies it is charged when an invoice is issued, and we will not back-bill earlier usage.