AI Models

Anthropic's most capable model. Exceptional at complex reasoning, multi-step coding, research synthesis, and long-context analysis up to 1M tokens. Best choice for demanding agentic workflows.

The ideal balance of capability and speed. Handles most coding, writing, and analysis tasks with near-Opus quality at significantly lower cost. Supports extended thinking and 1M context.

Previous generation Opus for high-demand workflows. Excellent reasoning at a proven-capability level. 200K context window. No thinking mode.

Fastest Claude model. Built for low-latency applications, classification, summarisation, and high-volume tasks where speed matters more than maximal capability.

OpenAI's flagship multimodal model. Handles complex reasoning, code generation, image analysis, and long-context tasks. The most capable GPT-5 variant, optimized for production use.

Coding-optimized GPT-5 variant. Superior at code generation, refactoring, test writing, and agentic development tasks. Ideal for Cursor AI and Claude Code-style workflows.

Balanced capability and cost. Strong at code, writing, and analysis. Available with $90/day or unlimited spend limits — ideal for sustained development workflows at controlled cost.

Google's most advanced Gemini model. Outstanding at multimodal reasoning, long-document analysis, and tasks requiring broad world knowledge. Native multimodal from ground up.

Fast, cost-efficient Gemini model for high-throughput pipelines. Excellent at classification, summarisation, structured output, and multi-turn conversation at scale.

State-of-the-art open-source model with frontier-level coding and math performance at a fraction of the price. Excellent for developers running high-volume code generation, data analysis, or research pipelines.