Bring your own model.

Anthropic is the maker of Claude, the model family Empryo defaults to and the one its own benchmarks are run on. It's a first-class built-in provider: one `--set-key` and every Claude model is available for chat, code, search, and every task-router slot. If you only set up one provider with Empryo, this is the one to pick.

OpenAI is one of Empryo's 21 built-in providers. Set one API key and Empryo's GPT-5 series models drive every part of the agent: AST-level edits through the editor, the ranked Genome graph, and per-slot routing. Reasoning effort and thinking budgets are wired in, so you control how hard each model thinks per task.

Empryo ships Google Gemini as a built-in provider. Set one API key and route any Gemini model — from the million-token Pro tiers to the cheap, fast Flash models — to whichever agent slot you choose. Gemini's large context and low Flash pricing make it the default workhorse for context compaction and read-only exploration inside Empryo.

OpenRouter is a built-in Empryo provider: one API key unlocks 200+ models from every major lab behind a single OpenAI-compatible endpoint. Set `openrouter` once and switch between Claude, GPT, Gemini, Grok, and open-weight models from inside Empryo without juggling separate accounts or keys.

Groq runs open-weight models on its custom LPU hardware at speeds no GPU host matches, and Empryo ships it as a built-in provider. Point Empryo at Groq for the parts of the agent loop where latency dominates: search, summaries, cleanup, and routing. Set one key and the whole model lineup is available.

Run open-weight models entirely on your own machine and point Empryo at them. Ollama is one of Empryo's 21 built-in providers, auto-detected at `localhost:11434` with no API key and no cloud round-trips. Pair a local model with Empryo's Genome graph and AST editor for fully offline, zero-cost agentic coding.

Already pay for GitHub Copilot? Empryo can drive its models from your existing subscription, no separate API bill. You get Empryo's graph-aware, AST-based agent loop running on Claude, GPT, and Gemini that are already included in your Copilot seat. It is an unofficial integration over the same internal API the IDE extensions use, so treat it as a convenience, not a supported contract.

Empryo treats "custom" as the universal escape hatch: any API that speaks the OpenAI `/v1/chat/completions` format becomes a first-class provider with one block of JSON, no plugin and no code. Use it to wire up Together, Fireworks, Cerebras, DashScope, a corporate gateway, or a local vLLM/llama.cpp server, then drive it with Empryo's Genome, AST editing, and task router exactly like a built-in.