Google Gemini

What it is

Gemini is Google DeepMind’s multimodal model family (text, image, audio, video) with very long context windows, accessible via the Gemini API and Vertex AI.

Natively multimodal models with the longest mainstream context windows.

When it fits — and when it doesn’t

Known weaknesses

Each one links to the public issue, changelog or docs page it came from, where we have one.

1. 01

   Free tier users' content is used to train and improve Google's products.

   The pricing documentation explicitly states that content used on the free tier is utilized to enhance product quality, a constraint that does not apply to paid or enterprise tiers.

   Source: ai.google.dev/pricing ↗
2. 02

   Grounding with Google Search is unavailable on the free tier.

   Developers using the free package cannot access Google Search grounding, which is a feature restricted to paid accounts.

   Source: ai.google.dev/pricing ↗
3. 03

   Free tier access is limited to specific models.

   The free tier provides restricted access to certain models, unlike paid tiers which offer access to Google's most advanced models.

   Source: ai.google.dev/pricing ↗

What it does well

• Very long context windows

  Million-token context for large documents/codebases

• Natively multimodal

  Text, image, audio, and video in one model family

• Competitive pricing (Flash)

  Low-cost fast tier for high volume

Pricing

Free tier

YesFree tier on Google AI Studio with rate limits

Paid from

$0.075/1M input tokens (Flash)

Model

Usage-based

Enterprise plan

Yes

Pricing not independently confirmed — check the vendor’s page.