Skip to content
LLMTR Docs

Frequently Asked Questions

What is LLMTR?

Section titled “What is LLMTR?”

LLMTR is a gateway platform that lets developers call Turkey-hosted models and global provider models through one OpenAI-compatible API. Model identifiers use the provider/model format, and one API key can access active public models in the catalog.

Is it compatible with the OpenAI SDK?

Section titled “Is it compatible with the OpenAI SDK?”

Yes. In existing OpenAI Python or JavaScript SDK usage, set base_url to https://llmtr.com/v1 and use your LLMTR API key.

Terminal window
curl https://llmtr.com/v1/chat/completions \
  -H "Authorization: Bearer llmtr-your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Is my data processed in Turkey?

Section titled “Is my data processed in Turkey?”

Models clearly marked as hosted in Turkey run on LLMTR infrastructure located in Turkey. When a global provider model is selected, request content may be processed on that provider’s infrastructure. Review each model page for the relevant data path and provider behavior.

Are prompts and responses stored?

Section titled “Are prompts and responses stored?”

LLMTR states data handling expectations on model pages and formal legal notices. Before sending sensitive content, review the selected model’s data path and provider terms.

How does pricing work?

Section titled “How does pricing work?”

Model usage costs are calculated from catalog model prices. LLMTR’s 6% platform margin is not added to model prices; it applies only to the total amount charged during credit top-up.

Which models are available?

Section titled “Which models are available?”

The active catalog may include models from OpenAI, Anthropic, Google, Mistral, xAI, Qwen, LLMTR, and other providers. Use the public model catalog for the current list.

Are streaming and multimodal requests supported?

Section titled “Are streaming and multimodal requests supported?”

Supported models can use streaming, image input, audio input or output, file flows, and image generation operations. Each model page lists its supported operations and modalities.

How are API keys protected?

Section titled “How are API keys protected?”

The raw API key is shown only once at creation time. Store it securely, keep it out of client-side code, and rotate it regularly.

What should be checked before production use?

Section titled “What should be checked before production use?”

Review the selected model’s data-processing geography, pricing unit, supported endpoint, rate-limit behavior, and fit with your regulated-data policy. For sensitive or regulated data, separately evaluate models hosted in Turkey.