Free AI chat aggregator that keeps your history out of the cloud

What it does Ptero is a web-based chat interface that lets you switch between models like DeepSeek V4, Llama 3.3, Qwen, MiniMax, and Mistral without creating an account. Everything runs in the browser, with conversations cached locally rather than on any server. The project also ships as a WordPress plugin, so you can drop the same interface into a site with a shortcode.

The interesting bit The non-profit project treats privacy as a default feature, not a premium upsell: there are no paid tiers, no subscriptions, and no server-side chat logs. That model is increasingly rare among AI frontends.

Key highlights

• Aggregates multiple models (DeepSeek, Llama, Qwen, MiniMax, Mistral) through one UI
• Zero signup; display name stored only on the local device
• All conversation history stays in the browser
• GPL v2 licensed and positioned as fully open-source
• Includes a WordPress plugin variant for self-hosted embedding

Caveats

• The README never explains how the project secures or funds API access to commercial models, so the “completely free” claim lacks visible mechanics.
• The WordPress plugin lives in a separate, unexplained repository (Pterocos), which makes the self-hosting story slightly confusing.
• The technology stack is described broadly (PHP REST API) but the architecture and data flow between browser, backend, and model providers remain unspecified.

Verdict Privacy-curious developers and WordPress admins who want a zero-cost, zero-signup AI chat UI will find Ptero useful. Anyone needing clarity on backend infrastructure, API rate limits, or long-term maintenance should treat the README as a starting point, not a specification.