Gemini
Multimodal assistant wired directly into Search, Workspace and Android.
What it is
Gemini wins on where it lives rather than what it does in isolation. If your files are already in Drive and your mail in Gmail, an assistant that reads both without an export step is worth more than a slightly better model you have to copy-paste into.
Pricing updated 2026-11-12 Official pricing
Gemini pricing explained
What you are buying is quota and access to the newest model, not a different assistant: the free tier gives you the same interface with tighter limits, the mid tier raises the limits and unlocks the newer models, and the top tier adds the largest allowance plus early access. The one thing to check before subscribing is whether the ecosystem lock-in is acceptable — the paid plan is only worth it if you actually use Drive, Gmail or Android, because outside Google it is a competent assistant without a structural advantage. Tier names and prices on the official page have changed several times; confirm the current numbers on one.google.com rather than trusting a review.
Free-tier limits are taken from the official pricing page on the update date above. They change often — confirm before you commit to an annual plan.
Gemini key features
Workspace integration
Reads Gmail, Drive and Calendar directly, so a question about last quarter can be answered from the actual documents instead of from a pasted summary.
Deep Research reports
Runs a multi-step search, reads the sources and returns a structured report with links. It is the closest thing here to handing off a literature scan.
Image and video understanding
Takes video and audio as input, not just stills — the only assistant in this group that reads a screen recording and answers questions about what happened in it.
Live voice conversation
Hands-free back-and-forth with interruption support. Useful while driving or cooking, and noticeably better at turn-taking than push-to-talk voice modes.
Gems — custom assistants
Saved instructions and files that preconfigure the model for one recurring job. The value is consistency across a team, not a new capability.
Google Search grounding
Answers that touch current facts are checked against Search rather than generated from memory, and the supporting links are attached.
Gemini: how it works
Multimodal from pretraining, not bolted on
The model is trained on text, images, audio and video together rather than having a vision model attached afterwards. That is why video questions work without a separate transcription step, and why the failure mode is different: it can misread a chart rather than ignore it.
Retrieval through Workspace permissions
When it reads your mail or files, it does so under your own account permissions, not a mirrored copy. The trade-off is honest: convenience inside Google, and very little of it outside.
Search grounding as a pipeline
For factual questions the model issues searches, reads results, then answers with citations. Answers are slower and more conservative, which is the intended trade when the alternative is a confident guess.
Long context with video
The window accepts hours of audio and video alongside text, which changes what "summarise this meeting" means — the model can attend to the segment, not just a transcript.
Gemini core strengths
Reads your actual workspace
Drive, Gmail and Calendar are readable without an export step. If your work already lives in Google, this removes the most tedious part of using an assistant.
The most usable free tier of the big three
Daily limits are generous enough to evaluate it properly, which matters if you are deciding between assistants rather than already committed.
Strongest media understanding
Video and audio in, answers out. For anything that involves watching or listening before answering, it is the one that does not need a preprocessing step.
Grounded answers with links
Factual claims arrive with sources you can open. That is not the same as being correct, but it is checkable, which is the property that matters for research.
Gemini compared with alternatives
| Tool | Gemini | ChatGPT | Claude | Grok |
|---|---|---|---|---|
| Best for | People already living in Workspace, Search and Android | Mixed daily work: writing, search, files, images | Long documents, and writing that survives several revisions | Live, breaking, "what is happening right now" questions |
| Free tier | The most generous free tier of this group | Usable but capped; newest models restricted | Daily message cap; uploads limited | Free tier with daily limits |
| Input and output | Text, image, audio and video in; image out | Text, image and voice in; text, image and voice out | Text and image in; text out, no image generation | Text, image and voice in; image out |
| Ecosystem and integration | Gems, Workspace integration, API — strongest inside Google apps | Custom GPTs, documented API, first-party desktop and mobile | Projects, Artifacts, API, MCP; smaller plugin ecosystem | Tied to the parent social platform; thin elsewhere |
| Paid entry price | $19.99 / month | $20 / month | $20 / month | Bundled with the platform subscription |
Written from the official product pages listed under Sources, not copied from another directory. Prices and limits reflect the update date above; features ship faster than reviews, so treat the list as a starting point and confirm on the official site.
Pros and cons
Strengths
- Reads Drive, Gmail and Calendar without exports
- Free tier is the most usable of the big three
- Strong image and video understanding
Limitations
- Answers vary more between runs
- Privacy settings take real effort to configure
- Best features assume a Google account
Gemini use cases
Email and calendar triage
Ask what needs a reply today and get an answer built from the actual threads, not from a summary you wrote yourself.
Video review
Upload a recorded call or demo and ask where the explanation lost the audience. It watches the segment instead of relying on a transcript.
Research with sources
Deep Research for a topic you know little about: it returns a structured report with links, which you can then verify line by line.
Document drafting inside Drive
Drafts land in Docs with the house formatting, because the integration writes into the file rather than into a chat box.
Android device assistance
System-level access on Android means it can act on what is on screen, which is the difference between describing a task and starting it.
