Poolside

Rho supports the Poolside API Platform through its OpenAI-compatible Chat Completions API.

Provider details

Setting	Value
Provider	poolside
Auth	poolside-api-key
Environment override	POOLSIDE_API_KEY
API base	https://inference.poolside.ai/v1
Model source	Poolside /models API

Create an API key in Poolside Platform. Poolside model access and model IDs depend on your account, so Rho fetches the model list after login instead of keeping a static list.

Interactive login

In the TUI, run:

/login poolside

Paste your Poolside API key when prompted. Rho stores it in the configured credential store and refreshes the models available to your account. You can then select one with /model, for example:

/model poolside/laguna-m.1

Rho stores the unprefixed model id (laguna-m.1) in cache and config. User-facing selection still uses poolside/laguna-m.1, and Rho sends the namespaced wire id poolside/laguna-m.1 to the Poolside API.

Remove the stored key with:

/logout poolside

Environment and automation

For CI or development, set POOLSIDE_API_KEY. It overrides a stored key:

export POOLSIDE_API_KEY="<api-key>"
rho --provider poolside \
  --auth poolside-api-key \
  --model poolside/laguna-m.1 \
  run "review this project"

Rho sends the key as a Bearer token. Do not put the key in config.toml or commit it to source control.

Models and reasoning

Use /config and choose Refresh model lists to fetch the current models for your Poolside account. Poolside reasoning models expose a binary thinking control. Rho shows Off and Max as the available reasoning levels:

• Off sends chat_template_kwargs.enable_thinking as false.
• Max leaves chat_template_kwargs unset, which uses Poolside's default thinking mode.

The models.dev catalog marks Poolside reasoning models with reasoning: true but does not advertise effort levels. Rho maps that Poolside-specific capability to Off and Max; it does not send OpenRouter's reasoning.effort field to the direct Poolside API.

See Poolside's API examples for request details.