Gemini · Data analysis
Gemini for data analysis
llmwise has 2 of Google's Gemini models, from Gemini 3.8 Flash (up to 250 messages a month on Pro) to Gemini 3.1 Pro (preview) (up to 125 messages a month). We ran the same data analysis prompts on every one and published every reply: which Gemini model to use, from the results, what each costs per message, and how to get more out of it.
Based on 10 of our test runs on , through OpenRouter with the app's own prompt and settings. Updated .
Short answer
In our test runs on September 27, 2026, Gemini 3.1 Pro (5 of 5) passed the most data analysis prompts: our pick for hard problems. For value, Gemini 3.8 Flash (4 of 5), 250 a month on Pro. It's also our pick for everyday.
Our picks for data analysis
Hard problems
Passed 5 of 5 data analysis prompts, both hard ones.
Best value
Passed 4 of 5 data analysis prompts, with 250 a month on Pro.
Everyday
Passed 4 of 5 data analysis prompts, at 250 a month on Pro. The same model as best value.
These picks aren't our opinion: they're what the results below give, by these rules, among the Gemini models in llmwise. They change when the results do.
- Hard problems: the model that passed the most prompts and, of those, the most hard ones. Models level on both share the pick: the prompts don't tell them apart, so we don't break the tie by price or by name.
- Best value: among the models that draw on the monthly allowance, the one with the most messages on Pro that passed no more than one prompt fewer than the top model. Ties go to the one that passed more, then to the lower cost per reply.
- Everyday: among the cheapest models on the page (the everyday models, which come from the daily count, when the page has any), the one that passed the most. Ties go to the one that passed more of the hard prompts, then to the lower cost per reply.
Our data analysis test runs, model by model
How each model did on our 5 data analysis prompts, what each reply counted as on Pro, and what it cost to run.
| Model | Passed | Hard ones | Messages used on Pro | Cost per reply | Time per reply |
|---|---|---|---|---|---|
| Gemini 3.1 Pro (preview)Google | 5 of 5 | 2 of 2 | 1 each, of 125 a month on Pro | $0.0155 | 9.9 s |
| Gemini 3.8 FlashGoogle | 4 of 5 | 1 of 2 | 1 each, of 250 a month on Pro | $0.0028 | 5.4 s |
One prompt, side by side
Each model's reply to “A median, filtered two ways”, as it came, with whether it passed and what it cost.
The prompt: A median, filtered two ways (hard)
Here are support tickets as CSV. What's the median hours_to_resolve for high-priority tickets handled by the Billing team? ticket_id,team,priority,hours_to_resolve T1,Billing,high,5.5 T2,Tech,high,12 T3,Billing,low,30 T4,Billing,high,2 T5,Tech,low,48 T6,Billing,high,9 T7,Billing,medium,16 T8,Tech,high,3.5 T9,Billing,high,7 T10,Billing,high,26 T11,Tech,medium,20 T12,Billing,high,4 Show your working briefly, then end with a line that says "Final answer: " followed by the answer alone.
Gemini 3.1 Pro
Passed: Final answer 6.25: right.
To find the median `hours_to_resolve` for high-priority tickets handled by the Billing team, we first filter the data for those specific tickets: * T1: 5.5 * T4: 2 * T6: 9 * T9: 7 * T10: 26 * T12: 4 Next, we sort these values in ascending order: 2, 4, 5.5, 7, 9, 26 Since there are 6 values (an even number), the median is the average of the two middle numbers (the 3rd and 4th values): (5.5 + 7) / 2 = 12.5 / 2 = 6.25 Final answer: 6.25
551 tokens in, 590 out (405 of them reasoning) · 7.1 s · $0.0082 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google ·
Gemini 3.8 Flash
Passed: Final answer 6.25: right.
1. Filter for tickets where `team` is "Billing" and `priority` is "high": - T1: 5.5 - T4: 2 - T6: 9 - T9: 7 - T10: 26 - T12: 4 2. Sort the values in ascending order: - 2, 4, 5.5, 7, 9, 26 3. Calculate the median of the 6 values (average of the 3rd and 4th values): - (5.5 + 7) / 2 = 12.5 / 2 = 6.25 Final answer: 6.25
551 tokens in, 169 out · 3.5 s · $0.0010 · 1 message on Pro · answered by google/gemini-3.8-flash via Google ·
The data analysis prompts, and how they're scored
Final answer. Automatic. The reply's last “Final answer:” line must hold the right value.
Each prompt was sent the way llmwise sends a message in a side-by-side comparison, which offers no tools: the app's own system prompt, the model's own settings, and Pro's reply size limit (8,000 tokens). Read all five data analysis prompts and how every reply was scored.
Prompts like these to try yourself
Gemini in llmwise
| Model | On Pro | On Free | Context window | Images | PDFs | Reasoning | API price per 1M, in / out |
|---|---|---|---|---|---|---|---|
| Gemini 3.1 Pro (preview)Google | 125/mo on Pro | Yes | 1.05M tokens | Yes | Whole file | Yes | $2.00 / $12.00 |
| Gemini 3.8 FlashGoogle | 250/mo on Pro | Yes | 1.05M tokens | Yes | Whole file | Yes | $0.75 / $3.75 |
Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.
What matters for data analysis
Calculations in code
Numbers done in a model's head are often slightly wrong. Analysis you can trust runs real code on the real data.
Reading PDFs and charts
Reports arrive as PDFs and charts as images. A model that takes the whole PDF file and reads images has more to go on.
Explaining the result
The value is in the explanation: what the numbers mean, what's uncertain, and what to check next.
Gemini for data analysis in llmwise
Bring the data in
Attach CSV files, PDFs of reports (up to 100 pages) and screenshots of charts.
Run Python on your data (paid plans)
On a paid plan the model can run Python on the files you attach, then hand you the charts or CSVs it saves. A run counts as 1 Claude Haiku 4.5 message.
Spreadsheets
Ask for a table and it opens as a spreadsheet you can edit and download as Excel or CSV.
The Analyst persona
Pick the Analyst persona for answers grounded in data, assumptions and confidence levels.
In llmwise, a message to Gemini goes to its maker, Google, or through OpenRouter when llmwise can't reach the maker directly. See the Privacy Policy.
Tips
Describe the columns and what one row means before asking questions.
Ask the model to show the code it ran, so you can check it.
Start with a sample; run the full file once the approach works.
Ask what could make the conclusion wrong.
Questions
Is Gemini good for data analysis?
In our test runs on September 27, 2026, Gemini 3.1 Pro (preview) passed 5 of 5, Gemini 3.8 Flash passed 4 of 5 of our data analysis prompts, against a best result of 5 of 5 among all 19 models. Every reply is published on this page and the methods page, so you can judge them yourself.
Which Gemini model should I use for data analysis?
Start with Gemini 3.8 Flash for cleaning data, pivots, standard statistics and writing the analysis code, and move up to Gemini 3.1 Pro (preview) for cleaning data, pivots, standard statistics and writing the analysis code.
Can I use Gemini for data analysis for free?
Yes: Gemini 3.1 Pro (preview) and Gemini 3.8 Flash are on the Free plan (a one-time trial of 5 messages).
Can it make charts?
On a paid plan, a code run can save a chart (or any file) and you download it from the chat. The model writes the plotting code and runs it once you approve.
Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.
See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.