Multi-provider support isn't just a feature — it's central to how squads work. Different models have different strengths, costs, and speed profiles. A squad can route its scanner to a fast, cheap model (Gemini Flash) for high-volume monitoring, its worker to a deep reasoning model (Claude Opus) for complex analysis, and its verifier to a mid-tier model for cost-effective quality checks.
Squads shells out to native AI CLIs. Each provider's CLI handles auth, context, and tool use independently — Squads just orchestrates.
| Provider | CLI | Status |
|---|---|---|
| Anthropic | claude |
Stable — primary provider |
gemini |
Stable | |
| DeepSeek | aider (delegation) |
New in 0.8.0 — file-based roles |
| OpenAI | codex |
Experimental |
| Mistral | vibe |
Experimental |
| xAI | grok |
Experimental |
| Ollama | ollama |
Experimental |
Experimental providers have CLI integration but haven't been extensively tested in production. Contributions welcome — especially from teams already using these CLIs for autonomous work.
squads run research --provider=google --model=gemini-2.5-flash
squads providers # List available providers and install statusSet the provider in an agent's frontmatter — mixed-provider squads work out of the box:
---
role: "scanner"
provider: "deepseek"
model: "deepseek-chat"
---Precedence: agent frontmatter → --provider flag → squad
providers.default.
Some providers (DeepSeek via aider) run as file-based executors: they
read and edit files and commit the result, but don't run shell commands
or browse the web. They're a fit for validators, formatters, and
summarizers — agents whose job is read files → write files. Squads
harvests their output from the isolated execution worktree, so completed
work is never lost even if the run ends abnormally.
Every provider run is recorded in observability (squads exec list,
squads usage) with real token and cost figures parsed from the
executor's output.