Comparison · Data analysis
ChatGPT vs Gemini for data analysis
On llmwise Pro, GPT-6 Sol gets up to 125 messages a month and Gemini 3.1 Pro (preview) up to 125 messages a month. ChatGPT is OpenAI's own app for its GPT models; llmwise has GPT models in its own chat, not ChatGPT itself. We ran the same 5 data analysis prompts on all 3 GPT models and all 2 Gemini models and published every reply: the results, then GPT-6 Luna against Gemini 3.1 Pro prompt by prompt, then how GPT and Gemini compare on price per message, context and files.
Based on 25 of our test runs on , through OpenRouter with the app's own prompt and settings.
Short answer
In our data analysis test runs on September 27, 2026, GPT's 3 models passed 15 of 15; GPT-6 Astra, GPT-6 Sol, and GPT-6 Luna each passed 5 of 5. Gemini's 2 models passed 9 of 10; its best, Gemini 3.1 Pro, passed 5 of 5 ($0.0155 a reply). GPT-6 Luna and Gemini 3.1 Pro each passed 5 of the 5 prompts, so these data analysis prompts don't split GPT and Gemini; the replies on this page show how they differ.
GPT and Gemini on our data analysis test runs
Every GPT and Gemini model in llmwise on our 5 data analysis prompts: how many replies passed, what each counted as on Pro, and what it cost to run.
| Model | Passed | Hard ones | Messages used on Pro | Cost per reply | Time per reply |
|---|---|---|---|---|---|
| GPT-6 AstraOpenAI | 5 of 5 | 2 of 2 | 1 each, of 31 a month on Pro | $0.0154 | 4.0 s |
| GPT-6 SolOpenAI | 5 of 5 | 2 of 2 | 1 each, of 125 a month on Pro | $0.0034 | 3.5 s |
| GPT-6 LunaOpenAI | 5 of 5 | 2 of 2 | 1 each, of 60 a day on Pro | $0.0002 | 3.4 s |
| 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 |
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.
GPT vs Gemini, prompt by prompt
Each data analysis prompt with GPT-6 Luna and Gemini 3.1 Pro side by side, as they replied: for each side, the model that passed the most data analysis prompts (then the most hard ones, then the cheaper).
Data analysis: GPT-6 Luna passed 5 of 5 and Gemini 3.1 Pro (preview) 5 of 5. GPT-6 Luna answered 3.5× sooner at the median, 2.5 s against 8.6 s. GPT-6 Luna cost 87.9× less, $0.0009 against $0.0773 for the 5 replies. Gemini 3.1 Pro (preview)'s replies ran 343% longer, in tokens of reply, thinking not counted.
1. The region with the most revenue
Everyday. GPT: 3 of 3 models passed, GPT-6 Luna the cheapest at $0.0001 · Gemini: 2 of 2 models passed, Gemini 3.8 Flash the cheapest at $0.0038. The prompt in full.
GPT-6 Luna and Gemini 3.1 Pro both passed this one. GPT-6 Luna answered sooner (2.5 s to 8.5 s), and GPT-6 Luna's reply cost less ($0.0001 to $0.0117); GPT-6 Luna wrote 70 tokens of reply, Gemini 3.1 Pro 255.
GPT-6 Luna
Passed: Final answer North — 1,085.00: right.
West: $90 + $310 + $240 + $225 = $865 East: $120 + $135 + $310 + $360 = $925 North: $180 + $240 + $620 + $45 = $1,085 Final answer: North — $1,085.00
640 tokens in, 135 out (65 of them reasoning) · 2.5 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
Gemini 3.1 Pro
Passed: Final answer North, 1085: right.
Here is the revenue calculation for each region (quantity × price): * **West**: * 2 × $45 = $90 * 1 × $310 = $310 * 2 × $120 = $240 * 5 × $45 = $225 * *Total* = $865 * **East**: * 1 × $120 = $120 * 3 × $45 = $135 * 1 × $310 = $310 * 3 × $120 = $360 * *Total* = $925 * **North**: * 4 × $45 = $180 * 2 × $120 = $240 * 2 × $310 = $620 * 1 × $45 = $45 * *Total* = $1085 Final answer: North, $1085770 tokens in, 849 out (594 of them reasoning) · 8.5 s · $0.0117 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google ·
2. Average order value in August
Everyday. GPT: 3 of 3 models passed, GPT-6 Luna the cheapest at $0.0001 · Gemini: 2 of 2 models passed, Gemini 3.8 Flash the cheapest at $0.0016. The prompt in full.
GPT-6 Luna and Gemini 3.1 Pro both passed this one. GPT-6 Luna answered sooner (2.6 s to 8.6 s), and GPT-6 Luna's reply cost less ($0.0001 to $0.0130); GPT-6 Luna wrote 64 tokens of reply, Gemini 3.1 Pro 273.
GPT-6 Luna
Passed: Final answer 300.00: right.
August revenue: $240 + $310 + $620 + $225 + $360 + $45 = $1,800. There were 6 orders, so the average is $1,800 ÷ 6 = $300.00. Final answer: $300.00
647 tokens in, 131 out (67 of them reasoning) · 2.6 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
Gemini 3.1 Pro
Passed: Final answer 300.00: right.
Here are the orders for August 2026 and their calculated revenue (quantity × price): * Order 1007: 2 × $120.00 = $240.00 * Order 1008: 1 × $310.00 = $310.00 * Order 1009: 2 × $310.00 = $620.00 * Order 1010: 5 × $45.00 = $225.00 * Order 1011: 3 × $120.00 = $360.00 * Order 1012: 1 × $45.00 = $45.00 Total revenue for August = $240 + $310 + $620 + $225 + $360 + $45 = $1800.00 Number of orders in August = 6 Average revenue per order = $1800.00 / 6 = $300.00 Final answer: 300.00
779 tokens in, 955 out (682 of them reasoning) · 8.6 s · $0.0130 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google ·
3. Revenue change from July to August
Everyday. GPT: 3 of 3 models passed, GPT-6 Luna the cheapest at $0.0001 · Gemini: 2 of 2 models passed, Gemini 3.8 Flash the cheapest at $0.0037. The prompt in full.
GPT-6 Luna and Gemini 3.1 Pro both passed this one. GPT-6 Luna answered sooner (2.5 s to 9.0 s), and GPT-6 Luna's reply cost less ($0.0001 to $0.0135); GPT-6 Luna wrote 66 tokens of reply, Gemini 3.1 Pro 356.
GPT-6 Luna
Passed: Final answer 67.4% increase: right.
July revenue: $1,075.00. August revenue: $1,800.00. Percentage change = (($1,800 − $1,075) ÷ $1,075) × 100 = 67.4% increase. Final answer: 67.4% increase
652 tokens in, 135 out (69 of them reasoning) · 2.5 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
Gemini 3.1 Pro
Passed: Final answer 67.4%: right.
**July 2026 Revenue:** * Order 1001: 2 x $45 = $90 * Order 1002: 1 x $120 = $120 * Order 1003: 1 x $310 = $310 * Order 1004: 4 x $45 = $180 * Order 1005: 3 x $45 = $135 * Order 1006: 2 x $120 = $240 * **July Total:** $1,075 **August 2026 Revenue:** * Order 1007: 2 x $120 = $240 * Order 1008: 1 x $310 = $310 * Order 1009: 2 x $310 = $620 * Order 1010: 5 x $45 = $225 * Order 1011: 3 x $120 = $360 * Order 1012: 1 x $45 = $45 * **August Total:** $1,800 **Percentage Change:** ((1800 - 1075) / 1075) * 100 = (725 / 1075) * 100 = 67.4418...% Final answer: 67.4%
786 tokens in, 990 out (634 of them reasoning) · 9.0 s · $0.0135 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google ·
4. A median, filtered two ways
Hard. GPT: 3 of 3 models passed, GPT-6 Luna the cheapest at $0.0001 · Gemini: 2 of 2 models passed, Gemini 3.8 Flash the cheapest at $0.0010. The prompt in full.
GPT-6 Luna and Gemini 3.1 Pro both passed this one. GPT-6 Luna answered sooner (2.5 s to 7.1 s), and GPT-6 Luna's reply cost less ($0.0001 to $0.0082); GPT-6 Luna wrote 75 tokens of reply, Gemini 3.1 Pro 185.
GPT-6 Luna
Passed: Final answer 6.25 hours: right.
Billing high-priority resolution times, sorted: 2, 4, 5.5, 7, 9, 26. With six values, the median is the average of the middle two: (5.5 + 7) / 2 = 6.25 hours. Final answer: 6.25 hours
512 tokens in, 128 out (53 of them reasoning) · 2.5 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
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 ·
5. Correlation between ad spend and sign-ups
Hard. GPT: 3 of 3 models passed, GPT-6 Luna the cheapest at $0.0004 · Gemini: 1 of 2 models passed, Gemini 3.1 Pro at $0.0310. The prompt in full.
GPT-6 Luna and Gemini 3.1 Pro both passed this one. GPT-6 Luna answered sooner (6.7 s to 16.5 s), and GPT-6 Luna's reply cost less ($0.0004 to $0.0310); GPT-6 Luna wrote 99 tokens of reply, Gemini 3.1 Pro 588.
GPT-6 Luna
Passed: Final answer 0.97: right.
Using \(r=\frac{\sum (x-\bar{x})(y-\bar{y})}{\sqrt{\sum(x-\bar{x})^2\sum(y-\bar{y})^2}}\), the centered sums are \(23{,}912.5\), \(624{,}687.5\), and \(971.5\). Thus \(r \approx 0.9707\). Final answer: 0.97451 tokens in, 651 out (552 of them reasoning) · 6.7 s · $0.0004 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
Gemini 3.1 Pro
Passed: Final answer 0.97: right.
To calculate the Pearson correlation coefficient ($r$), we use the formula: $r = \frac{n(\sum xy) - (\sum x)(\sum y)}{\sqrt{[n\sum x^2 - (\sum x)^2][n\sum y^2 - (\sum y)^2]}}$ Given the data for ad_spend ($x$) and signups ($y$): - $n = 8$ - $\sum x = 500 + 800 + 650 + 1200 + 900 + 300 + 1100 + 700 = 6,150$ - $\sum y = 42 + 55 + 49 + 71 + 60 + 35 + 64 + 58 = 434$ - $\sum xy = (500 \times 42) + \dots + (700 \times 58) = 357,550$ - $\sum x^2 = (500^2) + \dots + (700^2) = 5,352,500$ - $\sum y^2 = (42^2) + \dots + (58^2) = 24,516$ Plugging these into the formula: Numerator: $8(357,550) - (6,150)(434) = 2,860,400 - 2,669,100 = 191,300$ Denominator $x$: $8(5,352,500) - (6,150)^2 = 42,820,000 - 37,822,500 = 4,997,500$ Denominator $y$: $8(24,516) - (434)^2 = 196,128 - 188,356 = 7,772$ $r = \frac{191,300}{\sqrt{4,997,500 \times 7,772}}$ $r = \frac{191,300}{\sqrt{38,840,570,000}}$ $r = \frac{191,300}{197,080.11} \approx 0.97067$ Final answer: 0.97505 tokens in, 2,496 out (1,908 of them reasoning) · 16.5 s · $0.0310 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google ·
Each GPT model against each Gemini model
Every GPT model against every Gemini model on the same 5 data analysis prompts: passes, the median wait and what the replies cost. A lead under 10% counts as close.
GPT-6 Astra vs Gemini 3.1 Pro: 5 of 5 each; GPT-6 Astra answered 2.6× sooner at the median and cost about the same.
GPT-6 Astra vs Gemini 3.8 Flash: GPT-6 Astra 5 of 5, Gemini 3.8 Flash 4; GPT-6 Astra answered 1.6× sooner at the median and Gemini 3.8 Flash cost 5.6× less.
GPT-6 Sol vs Gemini 3.1 Pro: 5 of 5 each; GPT-6 Sol answered 3.1× sooner at the median and cost 4.6× less. GPT-6 Sol vs Gemini 3.1 Pro (preview), on every job.
GPT-6 Sol vs Gemini 3.8 Flash: GPT-6 Sol 5 of 5, Gemini 3.8 Flash 4; GPT-6 Sol answered 1.9× sooner at the median and Gemini 3.8 Flash cost 1.2× less. GPT-6 Sol vs Gemini 3.8 Flash, on every job.
GPT-6 Luna vs Gemini 3.1 Pro: 5 of 5 each; GPT-6 Luna answered 3.5× sooner at the median and cost 87.9× less.
GPT-6 Luna vs Gemini 3.8 Flash: GPT-6 Luna 5 of 5, Gemini 3.8 Flash 4; GPT-6 Luna answered 2.2× sooner at the median and cost 15.7× less. GPT-6 Luna vs Gemini 3.8 Flash, on every job.
How the data analysis replies are scored
Every GPT and Gemini reply above was checked the same way as every other model's, by the rules published with the prompts: how the data analysis prompts are scored, and each one in full.
The differences at a glance
What follows from each model's facts in our catalog.
The lineups
GPT: 3 models, GPT-6 Astra, GPT-6 Sol, and GPT-6 Luna. Gemini: 2 models, Gemini 3.1 Pro (preview) and Gemini 3.8 Flash.
Price per message
The least expensive GPT model is GPT-6 Luna (60 messages a day on Pro); the least expensive Gemini model is Gemini 3.8 Flash (250 messages a month on Pro).
Context window
GPT goes up to 1.05M tokens (GPT-6 Astra); Gemini up to 1.05M tokens (Gemini 3.1 Pro (preview)).
Images and PDFs
Every model here reads images. Every model here takes a PDF as a whole file.
On the Free plan
Free's one-time trial of 5 messages covers GPT-6 Sol, GPT-6 Luna, Gemini 3.1 Pro (preview), and Gemini 3.8 Flash. Paid plans have every model, with messages every month.
Every GPT and Gemini model's context window, files and API price: ChatGPT vs Gemini.
Where your messages go
In llmwise, a message to GPT or Gemini goes to the model's maker, or through OpenRouter when llmwise can't reach the maker directly. The Privacy Policy has the details.
Questions
Which is better, ChatGPT or Gemini for data analysis?
In our data analysis test runs on September 27, 2026, GPT's 3 models passed 15 of 15; GPT-6 Astra, GPT-6 Sol, and GPT-6 Luna each passed 5 of 5. Gemini's 2 models passed 9 of 10; its best, Gemini 3.1 Pro, passed 5 of 5 ($0.0155 a reply). GPT-6 Luna and Gemini 3.1 Pro each passed 5 of the 5 prompts, so these data analysis prompts don't split GPT and Gemini; the replies on this page show how they differ.
Is GPT or Gemini cheaper?
In llmwise, the least expensive GPT model is GPT-6 Luna (60 messages a day on Pro), and the least expensive Gemini model is Gemini 3.8 Flash (250 messages a month on Pro). At API list prices (September 2026), a typical message of 4,000 tokens in and 700 out costs $0.0008 on GPT-6 Luna and $0.0056 on Gemini 3.8 Flash.
Can I use GPT and Gemini in the same chat?
Yes. Ask GPT-6 Luna a question, then switch the picker to Gemini 3.1 Pro and ask again: Gemini 3.1 Pro sees the whole conversation, GPT-6 Luna's answer included.
Claude, GPT, Gemini, DeepSeek, Grok, Kimi, and GLM, 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.