Discover our Integrated AI Models

Anthropic's most capable Claude model.

Fast, cost-efficient Claude model for responsive tasks.

Latest Opus model for complex coding and professional work.

Fast, capable Claude model with strong coding and agentic performance.

Low-cost, million-token multimodal DeepSeek model.

Google's advanced reasoning model for complex multimodal work.

Low-cost, high-throughput multimodal Gemini model.

Fast Gemini model with strong reasoning and coding.

Google's newest fast multimodal Gemini model.

Google's open-weight Gemma 4 model on the Gemini API free tier.

Gemma 4 instruction model hosted by DeepInfra.

Current million-token GLM flagship hosted by DeepInfra.

Fast, low-cost multimodal GLM model.

OpenAI's flagship GPT-6 model for the hardest problems.

Fast, low-cost GPT-6 model for everyday tasks.

Balanced GPT-6 model for demanding professional work.

Large open-weight OpenAI model served at high speed by Groq.

Compact open-weight OpenAI model served by Groq.

Multi-agent Grok configuration for complex research tasks.

Low-latency configuration of the Grok 4.20 model family.

Reasoning configuration of the Grok 4.20 model family.

Million-token general-purpose Grok model.

xAI's latest flagship model for coding and chat.

Coding-focused Grok model for software development tasks.

Multimodal Kimi reasoning model hosted by DeepInfra.

Moonshot's largest multimodal reasoning model.

Large NVIDIA Nemotron reasoning model hosted by DeepInfra.

Efficient multimodal Qwen 3.6 mixture-of-experts model.

Preview multimodal Qwen reasoning model served by Groq.

Alibaba's flagship Qwen 3.8 model hosted by DeepInfra.