Skip to main content
Modellix aggregates production AI models and web tools from multiple upstream providers into one API, one billing account, and one Modellix API key. The service URLs below point to each provider’s official site or console for reference. You still call models and tools through Modellix—you do not need a separate key for every vendor.

Media Models

Async image, video, and speech at https://api.modellix.ai.

LLM

Sync chat gateway at https://llm.modellix.ai.

Tools

Sync Web Search and Web Fetch at https://tool.modellix.ai.
For help choosing a media model by quality, speed, or modality, see Model Selection. To browse media models by capability, use the category pages on modellix.ai. For LLM Model IDs and token rates, use Modellix LLM.

​
Media Models

Image, video, and speech generation use the async Media Model API. Submit a job, then poll Query Task Result.
These links open each provider’s own terms. Your use of Modellix remains covered by the Modellix terms. Most Alibaba media models run on Alibaba Cloud International, under the international product terms, privacy policy, and data processing addendum. Models without an international node use the China-region Bailian service agreement and the Aliyun privacy policy. The Bailian agreement contains the China-region data-processing rules. Kling, MiniMax, PixVerse, and Vidu have no separate public data processing addendum. The OpenAI row links Microsoft’s Azure terms, because those models are served through Azure OpenAI. The Google row links the Google Cloud Platform terms, privacy notice, and data processing addendum.

​
LLM

The LLM gateway is synchronous (optional SSE) at https://llm.modellix.ai. Pass model as a provider/name ID from the LLM catalog—for example openai/gpt-5.6-sol or anthropic/claude-sonnet-5. Protocols and client setup: LLM Overview. The table lists the upstream platforms Modellix routes those calls through. Cloud hosts (Azure, AWS Bedrock, GCP, Huawei Cloud, Tencent Cloud, OpenRouter) are not Model ID prefixes unless the catalog lists that prefix.
Qwen uses the Alibaba Cloud International product terms, privacy policy, and data processing addendum. Alibaba Media Models use that same international service when the model has an international node. ZAI, DeepSeek, and Moonshot use their China-region open platforms. Those China-region terms do not include a separate public data processing addendum. The OpenAI row is OpenAI’s own API. Azure is a separate host for OpenAI GPT models. The Anthropic row is the official Claude API. AWS Bedrock is a separate host for Claude models. Huawei Cloud and Tencent Cloud use their international platforms. The Huawei Cloud row links the international customer agreement, privacy statement, and data processing addendum. The Tencent Cloud row links the international terms of service, privacy policy, data processing and security agreement, and AI service terms.

​
Tools

Web Search and Web Fetch are synchronous calls at https://tool.modellix.ai. See Tools Overview for pricing, headers, and logs.
Tavily does not publish a separate data processing addendum. Customer-data rules are in the platform terms. Exa publishes a data processing addendum.

​
How Modellix Uses Provider Services

  • Unified API: Authenticate with your Modellix API key. Media Models use provider paths under https://api.modellix.ai (for example /alibaba/..., /google/...). LLM uses https://llm.modellix.ai with provider/name Model IDs. Tools use https://tool.modellix.ai.
  • Request model: Media and speech jobs are async—poll Query Task Result. LLM and Tools return synchronously (LLM may stream SSE).
  • Upstream URLs: The tables list vendor marketing or console sites so you can read provider-level product context. Each table also links the providers’ published legal terms. Runtime traffic for Modellix customers goes through Modellix infrastructure, not those URLs directly.

​
Availability

Modellix may add, retire, or forward providers and individual models based on product needs. Treat the Media Models navigation, LLM catalog, New Models, Retired Models, and Redirected Models as the source of truth for what is currently documented, which model IDs have been shut down, and which model IDs are forwarded to a newer version.