The models we build with and how we pick them.

For: hard decisions, document analysis, agent step planning

For: assistant on your knowledge, customer support, content translation

For: code generation and refactor, integrations, multilingual tasks

For: chat, quality comparisons, tasks with light reasoning

For: summaries, classification, extraction, cheap tasks at scale

For: writing code, automations, agent tooling, technical integrations

For: intent routing, lead classification, field extraction, quick answers

For: listing photo analysis, document reading, image description and tagging

For: deep reasoning, long contexts, quality comparisons in the arena

For: mixed tasks, a backup chat engine, arena experiments

For: real-time interfaces, fast feedback, short answers

For: fallback reasoning, analytical tasks, provider independence

For: fallback reasoning, high-volume analysis, quality comparisons

For: long-document analysis, whole knowledge bases in one pass

For: semantic search, RAG, data privacy — the foundation of any AI knowledge base

For: on-prem deployments, full control over weights, no cloud dependency

For: hardest reasoning, code work, quality-critical tasks

For: premium tasks where the client requires a specific provider, quality benchmarking

For: production assistants, code work, agents with a good quality-to-cost ratio

For: tasks needing current information, trend analysis, live research

For: enterprise-grade RAG, company search, assistants with hard citation

For: on-prem deployments on modest hardware, edge tasks, low unit cost