Contains affiliate links. We may earn a commission — it never changes the verdict.Details
The Stack / Chat Assistants
Google Gemini
02 DecisionsIf your operation lives in Google Workspace, Gemini is the highest-leverage assistant you can add — the integration does work the others physically can't. Outside Workspace it's a strong generalist but not a reason to switch stacks.
The job in the stack
The Google-side seat: massive context windows, Deep Research reports, and native reach into Gmail, Docs, and Drive.
Pricing
| Plan | Price | Notes |
|---|---|---|
| Free | $0 | Gemini 3.6 Flash default, daily 3.1 Pro allotment, 5 Deep Research reports/mo |
| Google AI Plus | $4.99/mo | $49.99/yr annual. Cut from $7.99 on June 8, 2026; storage doubled to 400 GB. 128K context, 2x free-tier limits |
| Google AI Pro | $19.99/mo | Gemini 3.1 Pro at full 1M context, 4x access, 5 TB storage |
| Google AI Ultra | $99.99/mo | 5x Pro limits, 20 TB storage. A $199.99 tier sits above it at 20x Pro limits — the old single $249.99 Ultra split into these two at I/O 2026 |
Pricing verified 2026-08-20. Plans change — the source of truth is Google Gemini's own pricing page.
Models & versions
- Gemini 3.7 FlashLaunched August 13-14, 2026, three weeks after 3.6 Flash — the new workhorse for coding and agent workloads. API pricing is $0.75/M input and $3.75/M output, and that is an introductory rate with a published expiry: it doubles to $1.50/$7.50 on January 1, 2027. Google moved 3.6 Flash onto the same rate, so the two cost the same today. In the consumer apps it is not the free default — 3.6 Flash still is; 3.7 Flash reaches consumers through Spark, Google's always-on agent, which needs AI Pro or Ultra
- Gemini 3.6 FlashReleased July 21, 2026 — the new free-tier workhorse, replacing 3.5 Flash: 17% fewer output tokens, output price cut $9.00→$7.50/M, coding and computer-use benchmarks both up, knowledge cutoff advanced to March 2026
- Gemini 3.5 Flash-LiteCheaper, faster tier for high-volume work, shipped alongside 3.6 Flash
- Gemini 3.1 ProFull 1M-token context on AI Pro plan
- Gemini 3.1 Deep ThinkExtended reasoning tier
- Nano Banana 2Image generation model; a faster Nano Banana 2 Lite shipped to the API July 2026
- Gemini OmniMultimodal video-generation line; Omni Flash in API public preview as of July 2026
The honest read
What works
- +1M-token context on AI Pro — reads what others can't hold
- +Deep Research produces genuinely usable source-linked reports
- +Native Workspace integration: Gmail, Docs, Sheets, Drive
- +Free tier is the most generous of the big three
What breaks
- −Model naming and plan structure change often enough to need a map
- −Writing voice tends flat compared to Claude for operator content
- −Best features assume you're all-in on Google's ecosystem
Use it for
- →Workspace-native teams and solo operators in Gmail/Docs all day
- →Research-heavy work needing huge context windows
- →Operators who want a capable free assistant
Skip it if
- ×Your files and mail live in Microsoft 365 — Copilot integration beats bolt-on Gemini
- ×You need the deepest writing/reasoning seat — that's Claude
Questions operators ask
?Is Google AI Pro worth $19.99?
If you're in Workspace daily, yes — the 1M context plus Gmail/Docs integration compounds. If you're not a Google-first operation, a Claude or ChatGPT subscription returns more.
?Is the $4.99 AI Plus tier enough?
For most operators, yes — as a second assistant. It buys 2x the free limits, a 128K context, and 400 GB of storage for less than a fifth of a Pro seat. What it does not buy is the 1M-token context, and that window is the whole reason to run Gemini next to Claude or ChatGPT. Plus if Gemini is your backup. Pro if Gemini is the seat doing the reading.
?What's the difference between Gemini and NotebookLM?
Gemini is the general assistant; NotebookLM is a grounded research notebook that only answers from sources you load. Different jobs — many stacks carry both.
What changed
- Correction: the free-tier line still named Gemini 3.5 Flash as the default model. 3.6 Flash took that seat on July 21, 2026 and still holds it — 3.7 Flash reaches consumers only through Spark, which requires AI Pro or Ultra. Consumer plan prices unchanged (Free / Plus $4.99 / Pro $19.99 / Ultra $99.99 and $199.99).
- News scan: Gemini 3.7 Flash launched August 13-14, 2026 and is the new default workhorse for coding and agentic work — added to the model list. Consumer plan prices (Free / Plus $4.99 / Pro $19.99 / Ultra $99.99 and $199.99) are unaffected and were not re-verified this pass. The detail worth planning around is on the API side: $0.75/M input and $3.75/M output is an introductory rate that doubles to $1.50/$7.50 on January 1, 2027. If you are costing an agent build on 3.7 Flash, cost it at both numbers.
- Correction — our Plus row was stale by two months. Google AI Plus dropped from $7.99 to $4.99/mo ($49.99/yr) on June 8, 2026, and storage doubled from 200 GB to 400 GB. Also corrected the top of the ladder: Ultra is now two tiers, $99.99 (5x Pro limits, 20 TB) and $199.99 (20x Pro limits), after the old $249.99 plan split at I/O 2026. Pro holds at $19.99 with 5 TB and 4x access.
- News scan: Google shipped Gemini 3.6 Flash (new free-tier default) plus 3.5 Flash-Lite on July 21 — updated models_or_versions; consumer plan pricing (Free/Plus/Pro/Ultra) unaffected.
- Verified: plan ladder unchanged (Free / Plus $7.99 / Pro $19.99 / Ultra $99.99). July additions: Nano Banana 2 Lite (fastest image model, API), Gemini Omni Flash in API public preview for video workflows, and Gemini Live folding into chat for seamless voice-text switching.
- Verified plan ladder: Free, AI Plus $7.99, AI Pro $19.99, AI Ultra $99.99. Gemini 3 family current: 3.5 Flash, 3.1 Pro, 3.1 Deep Think.
Give Google Gemini a job
The Google-side seat: massive context windows, Deep Research reports, and native reach into Gmail, Docs, and Drive.
Try Google Gemini →Alternatives with different jobs
Claude
02 DecisionsThe thinking seat: long documents, hard reasoning, code, and anything where the words have to be right the first time.
ChatGPT
02 DecisionsThe generalist seat: quick answers, brainstorming, voice mode, and the widest plugin/app ecosystem when you need breadth over depth.
Gemini Notebook (was NotebookLM)
01 IntelligenceThe source-grounded seat: synthesizing your documents, transcripts, and research into answers that cite themselves.