Gemini app
Good for writing on AI Pro with 3.8 Flash; weak at coding and hard questions, where Claude and ChatGPT score higher.
Going away: Gems in the Gemini app (moved into Skills), no date given yet. What Google offers instead: Skills. Source: gemini.google, read 19 Oct 2026.
Should you use it?
Pick it if
People who already pay Google for storage and want a writing helper in the same plan.
Best for: People who already pay Google for storage and want a writing helper (AI apps Top 20)
Skip it if
You want the best free help for coding or hard questions: Claude's free plan scores 80 for both; ChatGPT's free plan scores 60 for hard questions.
You give up: Free plan runs a much weaker model; usage limits are not published in numbers
The short version
- What this is
- Google's assistant app on web, Android, iPhone, Mac and Windows. Free lists Gemini 3.5 Flash-Lite; AI Pro adds 3.8 Flash and 3.1 Pro.
- Why it matters
- Readers compare it with ChatGPT and Claude; this shows which Google plan you need for its better models and where it trails.
- Free lists only 3.5 Flash-Lite, which ranks 85th of 364 on Text Expert s2 s5.
- AI Pro ($19.99/month) adds 3.8 Flash, 7th of 412 on Creative Writing, which earns it a score of 70 for writing s1 s5.
- Chats can train Google models unless you turn off Keep Activity s7.
Good at
Weak at
- The Free plan lists only 3.5 Flash-Lite, rank 85 of 364 on Text Expert and 74 of 142 on WebDev s2 s3 s5.
- Google does not say which model AI Pro uses by default, and its limits page names different models from its plans page s5 s8.
- Usage limits are published only as multiples of an unstated standard limit s8.
Our take
Gemini's scores depend on the plan. Google's plans page lists only Gemini 3.5 Flash-Lite on Free, and 3.8 Flash and 3.1 Pro on AI Pro at $19.99 a month, without naming a default s5. Our rule for every app is to score each plan on the best model it lets you choose, as the maker documents it. So AI Pro is scored on 3.8 Flash, which ranks above 3.1 Pro on all four boards we use s1 s2 s3 s4.
That gives AI Pro a score of 70 for writing: 3.8 Flash is 7th of 412 on Creative Writing s1. Coding, hard questions and agent work stay at 50, because 3.8 Flash ranks only 31st of 142 on WebDev, 17th of 364 on Text Expert and 22nd of 52 on the Agent board s2 s3 s4, and Free does no better. For coding and hard questions Claude's free plan scores 80. The best non-Anthropic picks we review are LM Studio, 80 for coding at $0, and Perplexity Pro, 70 for hard questions; ChatGPT scores 70 for coding and agents on Plus and 60 for hard questions on Free.
Google AI Ultra lists the same two models plus Deep Think, which has no board rank, and no Google plan page lists Gemini 4 Argon yet s5. The 3.8 Flash ranks are for its high setting, and Google says Standard thinking is the app's default s8.
Scores by job
A score from 0 to 100 per job, against everything available today. Overall 55 across 4 jobs, the average of the job scores.
New to this? What the scores and labels mean
- 0 to 100
- Our score for one job, like writing or coding, compared with everything else out today. 100 is the best available today. For AI models we build it from public data, such as Arena's blind votes and Epoch AI's independent tests; apps and tools are scored on our own checklist. The same tool can score differently for different jobs, and where two likely ranges overlap, we call it a tie.
- Our verdict
- Use it: a top choice for most people. Worth trying: good, and worth a go. Depends: good for some people or one kind of job. Wait: too early to tell, or not out for everyone yet. Skip: something else does the job better.
- First look
- Just released. We say what it is and whether to bother, and give it a score once public results exist.
- No score yet
- None of the public sources we use covers it yet, or the only evidence is from a source whose numbers we can't use, so we describe it in words instead.
- Best buy
- Nothing cheaper scores as well for that job.
- Best free
- The best score for that job that costs nothing, through a free plan or by running on your own computer.
- Pay as you go
- No monthly plan: you pay for each use, usually through a developer account.
- Checked
- The day we last confirmed the score and the price against the sources. Prices change often, so check the date.
- AI model
- The engine underneath, like GPT or Claude. You use it through an app or, if you code, directly.
- Open source
- Free to download and run on your own computer, so nothing leaves your machine.
Our checklist score, going by a public leaderboard and the maker’s pages.
Our checklist
Early, few votes
Sources and notes
Public leaderboardMaker’s pagesFree (Gemini 3.5 Flash-Lite, the only model the plans page lists, s5): 5/10, does the job 1 (arena.ai Creative Writing rank 62 of 412, gemini-3.5-flash-lite, s1), underlying quality 0, value 2 ($0), privacy 1 s7, limits 1 s8. AI Pro (lists 3.8 Flash and 3.1 Pro with no default named, so the better one, 3.8 Flash, s5): 7/10, does the job 2 (rank 7 of 412, gemini-3.8-flash-high, preliminary, s1), underlying quality 2, value 1 ($19.99/month, s5), privacy 1, limits 1. Score from AI Pro, the higher total.
Our checklist score, going by a public leaderboard and the maker’s pages.
Our checklist
Early, few votes
Sources and notes
Public leaderboardMaker’s pagesFree (Gemini 3.5 Flash-Lite, s5): 5/10, does the job 1 (arena.ai Text Expert rank 85 of 364, gemini-3.5-flash-lite, s2), underlying quality 0, value 2 ($0), privacy 1 s7, limits 1 s8. AI Pro (3.8 Flash, the better of its two listed models, s5): 5/10, does the job 1 (rank 17 of 364, gemini-3.8-flash-high, preliminary, s2), underlying quality 1, value 1 ($19.99/month), privacy 1, limits 1. A tie, so the score goes to the cheaper plan, Free.
Our checklist score, going by a public leaderboard and the maker’s pages.
Our checklist
Early, few votes
Sources and notes
Public leaderboardMaker’s pagesFree (Gemini 3.5 Flash-Lite, s5): 5/10, does the job 1 (arena.ai WebDev rank 74 of 142, gemini-3.5-flash-lite, 242 votes, s3), underlying quality 0, value 2 ($0), privacy 1 s7, limits 1 s8. AI Pro (3.8 Flash, the better of its two listed models, s5): 5/10, does the job 1 (rank 31 of 142, gemini-3.8-flash-high, preliminary, s3), underlying quality 1, value 1 ($19.99/month), privacy 1, limits 1. A tie, so the score goes to the cheaper plan, Free.
Our checklist score, going by a public leaderboard and the maker’s pages.
Our checklist
1 source so far
Sources and notes
Public leaderboardMaker’s pagesFree: not scored; the plans page lists agent tools such as Jules and Chrome auto browse only on paid plans, and 3.5 Flash-Lite is not on the arena.ai Agent board s4 s5. AI Pro (3.8 Flash, the better of its two listed models; agent features listed, s5): 5/10, does the job 1 (Gemini 3.8 Flash (High) arena.ai Agent rank 22 of 52, +2.75%, s4), underlying quality 1, value 1 ($19.99/month), privacy 1 s7, limits 1 s8. Score from AI Pro.
Alternatives
Better or cheaper for writing, from our scores, then the ones we'd also look at. The same kind of product comes first.
- ClaudeApp70$20 a monthFree plan scores 80 for coding and hard questions, against 50 here. Side by side
- ChatGPT (GPT-6)App50$20 a monthBest all-round non-Anthropic assistant we review for these jobs: 70 for coding and agents on Plus, 60 for hard questions on Free. Side by side
- LM Studio (with Bionic agent)App80CodingFreeAlso scores 80 for coding at $0, running open models on your own computer, though those models rank lower. Side by side
- Claude Opus 5.5BetterText model100$20 a monthScores 100 in writing, against 70 here; $20 a month. Side by side
- Gemini 4 ArgonCheaperText model99$4.99 a monthScores 99 in writing for $4.99 a month, against $19.99 a month here. Side by side
Value for money
Where it sits in writing: higher is a better score, further left is cheaper. Green dots are the Best buys.
Evidence
Everything behind the call and the scores, folded so the page stays short. Spotted a wrong price, score or link? Report a mistake.
How we scored this
The writing, research, coding and agents scores come from our own checklist of the product.
Public leaderboardMaker’s pages
Reviews are drafted with help from AI models, including Anthropic's Claude. Model scores come only from outside leaderboards and independent evaluations, by a written rule.
No one paid for this review or gave us free access, and we earn nothing if you sign up. How we score
Sources 9
- s1arena.ai Text: Creative Writing leaderboard
- s2arena.ai Text: Expert leaderboard
- s3arena.ai Code Arena: WebDev leaderboard
- s4arena.ai Agent leaderboard
- s5Google AI plans (Gemini subscriptions)
- s7Gemini Apps Privacy Hub
- s8Gemini Apps Help: usage limits
- s9Engadget: A native Gemini app is finally available for Windows PCs (10 Sep 2026)
- s10Android Authority: Gemini's new Windows app (10 Sep 2026)
Something here out of date or wrong? Report a mistake and we will check it against the source.
History 4 changes
Every change to a rating, the price or our call, with the reason. Scores are against what exists on the day, so they drift down as better things ship. Older entries come from before we moved to scores.
- New: Writing ratedNew: Research ratedNew: Coding ratedNew: Agents ratedReviewed: writing, research, coding, agents.
- Better: Writing upVerdict Skip to DependsAudit under our rule that each plan is rated on the best model it includes: Google names no default for AI Pro, so AI Pro is now rated on its better model, 3.8 Flash, instead of the lower-ranked 3.1 Pro. Writing rose to 7/10. Research, coding and agents unchanged. Verdict now depends: it scores best for writing, where Claude Pro scores the same, so Gemini is not outclassed there.
- Rule change: best model each plan lets you choose. AI Pro still offers 3.8 Flash and 3.1 Pro, and Free only 3.5 Flash-Lite, so no rating changes. Arena text URLs moved to the /leaderboard/chat/text/ form.
- Like-with-like check: ratings compare Free with AI Pro only. Google's current plans page (fetched with a cache-buster) lists only 3.5 Flash-Lite on Free, so the Free doubt line and the release-notes source were removed. Ultra adds no ranked model, and no plan page lists Gemini 4 Argon. No rating changes.
Use it well 4 guides on this site
Pages on this site that help with the jobs above.