Models

Latest R1 update, near o3-level reasoning performance. Free on Waterfall.

Latest Llama with 1M context and excellent tool calling. Free on Waterfall.

DeepSeek R1 + web search. Reasoning with real-time data.

Anthropic's latest balanced model. Excellent coding and reasoning.

Excellent tool calling model on Groq free tier. 1000 req/day free.

State-of-the-art reasoning model competitive with o1. Completely free through Waterfall.

Google's most capable model with 1M context. Free experimental access through Waterfall.

Fast Llama 4 variant with 512K context. Free with tool calling support.

Qwen's reasoning model. Competitive with DeepSeek R1 at 32B parameters. Free.

Open-weights model from Google. Great general purpose performance. Free.

Excellent all-around chat model. 128K context. Free on Waterfall.

Mistral's best small model. Tool calling, 128K context. Free.

Fast multimodal model with vision and tool use. 1M context. Free.

NVIDIA-optimized Llama 3.1. Very fast, 128K context. Free.

Llama 4 Scout on Groq. Ultra-fast inference at $0.11/M input tokens.

Full DeepSeek V3. Excellent quality at extremely low cost.

Fast and capable. Great for high-volume use cases.

OpenAI's small model. Fast, cheap, reliable tool calling.

Quick Q&A with web search and citations. Real-time information.

Deeper analysis with web search. Multi-step research capability.

OpenAI's flagship multimodal model. Excellent all-around performance.

Google's best model. 1M context with multimodal capabilities.

Full R1 reasoning model. o1-level performance at a fraction of the cost.

OpenAI's reasoning model. Excels at math, science, and complex logic.

Anthropic's most capable model. Best-in-class for complex tasks.

Qwen3 on Groq free tier. Tool calling support, 1000 req/day.

Proven reliable Llama model. 128K context with tool calling. Free.