Provider setup
Connect your own application
Create credentials at your provider, configure them on your own machine, and make a small local verification request. This site never asks for, stores or tests your credentials.
Anthropic
Check the Claude Console billing page and available API credits. Access and limits are specific to your organization and workspace.
Researched from Anthropic's official documentation on 2026-11-02. Not yet walked through end to end.
Azure OpenAI
Requires an Azure subscription, a compatible model deployment and quota. Prices vary by deployed model and region; check the resource’s Azure pricing before calling it.
Researched from Azure OpenAI's official documentation on 2026-11-02. Not yet walked through end to end.
DeepSeek
API calls use your API platform balance and model-specific rates. An empty balance is a separate failure from an invalid API key; review the platform billing before sending a request.
Researched from DeepSeek's official documentation on 2026-11-02. Not yet walked through end to end.
ElevenLabs
Speech synthesis consumes your plan’s character or credit allowance. Check the selected voice/model price, remaining quota and per-key credit limit. The local example synthesizes only “Hello.”
Researched from ElevenLabs's official documentation on 2026-11-02. Not yet walked through end to end.
Gemini
Some models have limited free quotas. Paid API usage requires the project’s current billing setup; check AI Studio before sending a request.
Researched from Gemini's official documentation on 2026-11-02. Not yet walked through end to end.
Groq
Groq has model-specific free limits and paid plans. Review the project’s quota, billing and spend limits before using a paid tier.
Researched from Groq's official documentation on 2026-11-02. Not yet walked through end to end.
Mistral
Studio documents a limited Free mode without a credit card. Check the organization’s plan and quotas; activate billing for paid usage.
Researched from Mistral's official documentation on 2026-11-02. Not yet walked through end to end.
Ollama
Local models need your own hardware and electricity, but no provider API key or per-token cloud bill. Download size and memory requirements depend on your model. Cloud models are a separate billed service.
Researched from Ollama's official documentation on 2026-11-02. Not yet walked through end to end.
OpenAI
Check your API organization’s credits and billing before making a paid request. Model access and rate limits depend on your organization.
Researched from OpenAI's official documentation on 2026-11-02. Not yet walked through end to end.
OpenRouter
Model prices, provider charges and free-model limits vary. Review your available credits and key limits. A free model or a key does not imply unlimited free access.
Researched from OpenRouter's official documentation on 2026-11-02. Not yet walked through end to end.
Perplexity
Perplexity API usage is pay as you go. Check project credits and the selected API/model’s token, search or request fees; a consumer subscription does not establish API credits.
Researched from Perplexity's official documentation on 2026-11-02. Not yet walked through end to end.
Qwen / Model Studio
Model Studio bills by model and region. Verify the applicable quota, payment method and pay-as-you-go price before making a request.
Researched from Qwen / Model Studio's official documentation on 2026-11-02. Not yet walked through end to end.
Replicate
Review Replicate billing before running a model. The optional one-image verification example uses FLUX Schnell, approximately US$0.003 per output at the documented rate; actual charges and availability can change.
Researched from Replicate's official documentation on 2026-11-02. Not yet walked through end to end.
Requesty
Free models and paid routing have different limits. Check the selected model’s price, available credits and team/key spend controls before making a request.
Researched from Requesty's official documentation on 2026-11-02. Not yet walked through end to end.
Vertex AI (Google Cloud)
Requires a Cloud project with billing, the relevant API enabled, model access and regional quota. Token rates and other charges depend on your model and region.
Researched from Vertex AI (Google Cloud)'s official documentation on 2026-11-02. Not yet walked through end to end.
Voyage AI
Embeddings are billed by the selected model and input tokens, subject to your current allowance and plan. The local example embeds only “Hello.”; review the dashboard’s billing and limits first.
Researched from Voyage AI's official documentation on 2026-11-02. Not yet walked through end to end.
xAI
The xAI quickstart requires API credits. Check your team’s balance, model access and spend limits in the console.
Researched from xAI's official documentation on 2026-11-02. Not yet walked through end to end.
Z.AI
Check the standard API account’s balance and model pricing. The GLM Coding Plan uses a separate endpoint; do not assume that subscription covers the package’s default standard API endpoint.
Researched from Z.AI's official documentation on 2026-11-02. Not yet walked through end to end.