All tracked AI models

Anthropic's flagship. Deep reasoning, long-horizon agents.

Sonnet strikes back. The price-performance sweet spot.

Fast, cheap, surprisingly smart.

OpenAI's flagship. Omni-modal, omni-purpose.

GPT-5 shrunk for scale.

Long-context workhorse still on the menu.

Reasoning model. Thinks before it speaks.

OpenAI's open-weights return.

Google's flagship. Massive context, native multimodality.

Cheap, fast, 2M-token context.

On-device first. Free inference, private by default.

Real-time-aware, unfiltered, strong reasoner.

Open-weights heavyweight. Self-host or run anywhere.

Mid-size open model. Excellent price-performance when self-hosted.

Open-weights reasoning at bargain pricing.

European flagship with strong multilingual chops.

Small, open-weights, surprisingly sharp.

Alibaba's flagship. Top-tier multilingual and coding.

Open-weights coding specialist.

RAG-tuned enterprise workhorse.