Input capability

Free multimodal API candidates for vision or audio

This shortlist includes models carrying a Vision or Audio capability tag in the catalog. Open the model details before integrating because accepted formats and endpoint behavior vary.

Current shortlist1 active model candidates

Latest provider source check: 2026-07-28.

What this shortlist means

The inclusion rules are visible and intentionally narrow.

  1. 01

    Inclusion rule

    The model is tagged with Vision or Audio capability.

  2. 02

    Inclusion rule

    The associated provider profile is active and source-dated.

  3. 03

    Inclusion rule

    No text-only route is added merely because its provider offers other multimodal models.

Important limit: A modality tag does not prove that every listed endpoint, free quota or client supports the same file formats. Confirm the exact model documentation and test a non-sensitive sample.

Matching candidates

Live catalog results with provider evidence dates and the limits you should read before signup.

Model and providerWhy it matchesEvidence and limitsNext steps
Gemini Flash
Google Gemini
gemini-flash-latest
Vision + Audio
Account-specific no-card tier
Official sources: 2026-07-28
Not required to start the documented free tier
Project and model specific
Independent Canary pending