Claude vs ChatGPT vs Gemini : Which AI Assistant Is Best for You?
Home / Online BCA / Claude vs ChatGPT vs Gemini : Which AI Assistant Is Best for You?
Pick ChatGPT for the widest all-round assistant, native image generation and the cheapest paid entry ($8 Go). Pick Gemini if you live in Gmail, Docs and Android, want AI video, or want the best value bundle. Pick Claude for coding, long multi-step agent work and top benchmark intelligence. All three have usable free plans.
- Smartest model on independent tests: Claude Fable 5.2 leads the Artificial Analysis Intelligence Index at 65, ahead of Gemini 4 Carbon (60) and GPT-6 Astra (57).
- Most preferred in blind chat votes: Gemini 4 Carbon tops the LMArena Text leaderboard (1528).
- Best for coding: Claude. Opus 5.5 ranks first on LMArena’s WebDev board.
- Cheapest paid plan: Google AI Plus ($4.99/month) and ChatGPT Go ($8/month). In India both cost ₹399.
- Images and video: Gemini makes both. ChatGPT makes images only, since OpenAI shut down Sora in April 2026. Claude has no built-in image or video model.
- Standard plans cost the same: Claude Pro, ChatGPT Plus and Google AI Pro are all about $20 a month.
Our hands-on test (Nov 17, 2026): in 15 runs, all three got every checkable answer right. Gemini made images fastest, and only Gemini and Claude wrote for the Indian reader we specified.
Claude vs ChatGPT vs Gemini at a glance
At the $20 tier the three cost the same. They differ in which models you get, what media they can make, and which apps they plug into.
| Claude | ChatGPT | Gemini | |
|---|---|---|---|
| Made by | Anthropic | OpenAI | |
| Top model (Nov 2026) | Fable 5.2 | GPT-6 Astra | Gemini 4 Carbon |
| Main model on the ~$20 plan | Opus 5.5 and Sonnet 5.5 | GPT-6 Sol | Gemini 4 Argon |
| Free-plan model | Sonnet 5.5 | GPT-6 Luna | Gemini 4 Hydrogen |
| Cheapest paid plan (US) | Pro, $20/mo | Go, $8/mo | AI Plus, $4.99/mo |
| Top consumer plan (US) | Ultra, $200/mo | Pro, from $100/mo | AI Ultra, $199.99/mo |
| Image generation | Not built in; only through connected apps | Yes, Images 2.0 on all plans | Yes, Nano Banana |
| Video generation | No | No (Sora closed April 2026) | Yes, Veo on paid plans |
| Do-it-for-you agent | Cowork, built into every chat | ChatGPT Work | Gemini Spark |
| Coding tool | Claude Code | Codex | Antigravity |
| Context window (API) | 1M tokens | 1.05M tokens | 1.05M tokens |
| Chats used for training | You choose in settings | On by default, opt-out | On by default, opt-out |
| Best for | Coding, agents, deep reasoning | All-round use, images, app plugins | Google apps, video, value bundles |
Prices are US web prices before tax, checked against claude.com/pricing, gemini.google/plans and OpenAI’s pricing page.
What’s new in Claude, ChatGPT and Gemini in late 2026?
All three companies replaced their main models between September and November 2026. A comparison that still names GPT-5.x, Gemini 2.5 or 3, or Claude 4 describes products you can no longer buy.
| Date | Release | What it means for you |
|---|---|---|
| Nov 12, 2026 | Gemini 4 Hydrogen | Google’s fast Gemini 4 model; it powers the free and AI Plus tiers |
| Nov 10, 2026 | Gemini 4 Carbon | Google’s frontier model; in the Gemini app for AI Ultra subscribers |
| Nov 9, 2026 | Claude Fable 5.2 | Anthropic’s top model, on every paid Claude plan |
| Oct 31, 2026 | Claude Ultra plan | Max 20x renamed Ultra: $200, 10x Pro’s weekly usage, no five-hour limit for verified accounts |
| Oct 26, 2026 | GPT-6.1 Terra | Fast OpenAI model for computer use, in ChatGPT Work and Codex |
| Oct 13, 2026 | Gemini 4 Argon | Goes public for Google AI Pro and Ultra subscribers |
| Oct 7, 2026 | Claude Haiku 5.5 | Anthropic’s cheapest, fastest model |
| Oct 7–8, 2026 | GPT-6 in ChatGPT | GPT-6 Sol becomes the paid-plan default; Free and Go get GPT-6 Luna |
| Sep 28, 2026 | Claude Sonnet 5.5 | 30%+ faster than Sonnet 5 and up to 30% cheaper for most work |
| Sep 22, 2026 | Claude Opus 5.5 and GPT-6 Sol and Luna |
Opus 5.5 costs 40% less to run than Opus 5; Sol and Luna bring GPT-6 to cheaper tiers |
| Sep 16, 2026 | Cowork in every Claude chat | Claude’s file-and-app agent no longer needs a separate mode |
| Sep 3, 2026 | GPT-6 Astra | OpenAI’s new flagship and the first GPT-6 model |
One promised model never arrived. Google announced Gemini 3.5 Pro at I/O in May 2026 for a June launch, and it still had not shipped when Gemini 4 replaced it.
Read More: Claude vs ChatGPT: Which Is Better in 2026?
Which models power Claude, ChatGPT and Gemini?
Each company now sells a family of models, from a costly frontier model down to a cheap fast one. The name on the app is not the name of the model answering you, and your plan decides which one you get.
| Tier | Claude (Anthropic) | ChatGPT (OpenAI) | Gemini (Google) |
|---|---|---|---|
| Frontier | Fable 5.2 | GPT-6 Astra | Gemini 4 Carbon |
| Workhorse | Opus 5.5, Sonnet 5.5 | GPT-6.1 Sol | Gemini 4 Argon |
| Fast and cheap | Haiku 5.5 | GPT-6.1 Terra, GPT-6 Luna | Gemini 4 Hydrogen, Gemini 3.8 Flash |
| Workhorse API price, per 1M tokens in/out | Sonnet 5.5: $2 / $10; Opus 5.5: $4 / $20 | Sol: $2 / $10 | Argon: $2 / $10, rising to $4 / $20 on Dec 1, 2026 |
Claude. Anthropic’s names run Fable, Opus, Sonnet, Haiku, from most to least capable. Opus 5.5 has a 1M-token context window and 128K-token output. Fable 5.2 is included on Max and Ultra within plan limits; Pro users pay for it with usage credits. A stronger sibling, Mythos 5.2, stays restricted to vetted cyber and life-science work.
ChatGPT. OpenAI’s GPT-6 family runs Astra, Sol, Terra, Luna. In the regular chat window, Plus and higher plans get GPT-6 Sol and Free and Go get Luna, which OpenAI pitches for “high-volume tasks with a clear goal.” Terra, built for computer use, so far appears only in ChatGPT Work and Codex.
Gemini. Google names Gemini 4 models after elements. Carbon is the frontier model, Argon the one Google recommends for most knowledge work, and Hydrogen the fast one. In the Gemini app, AI Pro gets Argon and Ultra gets Carbon.
Which AI is smartest on independent benchmarks?
On the broadest independent score, Claude’s top model leads. On blind human votes, Gemini’s does. OpenAI’s best model sits close behind both.

The Artificial Analysis Intelligence Index combines tests of reasoning, knowledge, science and coding. Gemini’s free-plan model, Gemini 4 Hydrogen, scores 54, one point behind Claude Sonnet 5.5. Intelligence is not the only cost: Fable 5.2 at max effort costs about 3.4x as much per task as Gemini 4 Carbon. In September, GPT-6 Astra matched the previous leader at about 40% of its cost.
| Leaderboard (as of) | What it measures | #1 |
|---|---|---|
| LMArena Text (Oct 28, 2025) |
Blind human votes between two answers | Gemini 4 Carbon, 1528 |
| LMArena WebDev (Oct 17, 2026) |
Blind votes on web apps the models build | Claude Opus 5.5 |
| AA Coding Agent Index (Sept 2026) |
Multi-step coding tasks | Claude Sonnet 5.5, 68 |
One caution: leaderboards test the strongest model at its highest setting. On a $20 plan you mostly chat with GPT-6 Sol, Gemini 4 Argon or Claude Opus 5.5 and Sonnet 5.5, not the frontier models above.
We tested all three: 5 prompts, 15 runs
On November 17, 2026 we gave ChatGPT, Gemini and Claude the same five tasks. All three got every checkable answer right. They differed on images, speed and how well they wrote for an Indian reader.
Setup: ChatGPT Plus (default GPT-6 model, Instant mode), Gemini’s free plan (Auto) and Claude Sonnet 5.5 at High effort, the model on Claude’s free plan. Each prompt ran once, in a fresh chat, in Chrome from India.
| Task and how we scored it | ChatGPT | Gemini | Claude |
|---|---|---|---|
| Writing: 80–100-word product blurb for Indian students; no “!” and three banned words | Pass, 86 words plus a headline; never mentions India | Pass, 98 words; written for India (summer heat, chai, exam weeks) | Pass, 93 words; written for India (hostel, metro, canteen) |
| Coding: a second_largest() function, run against 14 test cases | 14 of 14 | 14 of 14, after the first attempt hung for 90 seconds | 14 of 14 |
| Live fact: RBI repo rate and when it last changed | Correct (5.50%, Oct 7, 2026); cited RBI | Correct; cited RBI and Bajaj Finance | Correct; cited RBI’s own resolution |
| Reasoning: when and where two trains meet | Correct: 7:40 AM, 133.33 km | Correct; ran code to check | Correct; showed its working |
| Image: a Mumbai tea stall at sunrise, watercolor | Made it in about 30 s | Made it in about 20 s | No built-in image model; made it through a connected image app in about 60 s |

We checked the code against our own 14 test cases, the repo rate against RBI’s October 7, 2026 decision, and word counts with a script.
What we learned
- Accuracy no longer separates them on everyday tasks. All 12 checkable answers were correct, and all three cited a source for the live fact.
- Only two wrote for the reader we named. Gemini and Claude localized the copy for Indian students; ChatGPT’s version could run anywhere.
- Watch for invented specs. Gemini and Claude both made up capacity and insulation figures. Only Claude flagged them as placeholders to replace.
- Gemini was the fastest image maker. Claude needed a connected third-party app, which took three times as long and used that app’s credits.
- One run is a snapshot, not a benchmark. Answers vary between runs, and the plans differed, so treat the speeds as indicative.
Every prompt and response, word for word, is in the appendix at the end of this article.
Claude vs ChatGPT vs Gemini pricing: how much does each cost?
The standard plans all cost about $20 a month, so the real differences sit at the bottom and the top. Gemini is cheapest to start, ChatGPT has the most tiers, and Claude has no budget plan.
US prices (per month, before tax)
| Level | Claude | ChatGPT | Gemini (Google AI) |
|---|---|---|---|
| Free | Free | Free | Free |
| Budget | No budget plan | Go, $8 | AI Plus, $4.99 |
| Standard | Pro, $20 ($17 billed yearly) | Plus, $20 | AI Pro, $19.99 |
| Power | Max, $100 | Pro, from $100 | AI Ultra, $99.99 |
| Top | Ultra, $200 | Higher Pro tiers | AI Ultra, $199.99 |
Sources: Claude plans, ChatGPT plans, Google AI plans.
India prices (per month, GST included)
| Level | Claude | ChatGPT | Gemini (Google AI) |
|---|---|---|---|
| Budget | No budget plan | Go, ₹399 | AI Plus, ₹399 |
| Standard | Pro, ₹2,399 (₹2,000 billed yearly) | Plus, ₹1,999 | AI Pro, ₹1,950 |
| Power | Max, from ₹11,999 | Pro, from ₹10,699 | AI Ultra, see Google’s India plan page |
Anthropic began billing Claude in rupees in July 2026; India is its second-largest market. Prices above were checked in November 2026. Jio Unlimited 5G users may still get 18 months of Google AI Pro free through the MyJio app.
What the price actually buys
- Gemini bundles the most. AI Pro includes 5 TB of Google storage and YouTube Premium Lite; AI Plus includes 400 GB.
- ChatGPT is ad-free from Plus up. OpenAI says Plus, Pro, Business and Enterprise carry no ads; Free and Go can show them.
- Claude’s tiers scale evenly. Max gives 5x Pro’s usage for 5x the price, and Ultra 10x for 10x. Max and Ultra also include $100 and $200 a month in API credits.
- Claude’s top model costs extra on Pro. Since the launch offer ended on November 16, 2026, Pro users pay for Fable 5.2 with usage credits.
Head-to-head: which is best for what?
Each verdict below names a winner and the evidence behind it. Rankings shift almost monthly, so run your own tasks on the free plans before you pay.
Which is best for writing and editing?
Verdict: close call, with Gemini ahead in blind votes. Gemini 4 models lead the LMArena Text leaderboard, where people pick between two anonymous answers. That board covers far more than writing, though. Claude keeps a strong reputation for following style rules without filler, and GPT-6 Sol is a capable all-rounder. Paste the same brief into all three and judge the drafts yourself.
Which is best for coding and software agents?
Verdict: Claude. Claude Opus 5.5 ranks first on LMArena’s WebDev board, and Sonnet 5.5 launched at #1 on Artificial Analysis’ Coding Agent Index. Anthropic’s system card puts Opus 5.5 at 89.9% on SWE-Bench Pro. OpenAI competes on cost: GPT-6 Astra matched Claude Fable 5.1 on that index at about 60% of the cost, and Codex has over 5 million weekly users. Google’s Antigravity is the third option.
Which is best for research and real-time information?
Verdict: Gemini for quick lookups; a tie for long reports. Gemini is built on Google Search, and its free plan includes Deep Research. ChatGPT and Claude also search the web and write multi-source research reports with citations.
Which makes the best images and video?
Verdict: Gemini, the only one that makes both. Its free plan includes Nano Banana image generation. Paid plans add Veo video: a Veo 3.1 Lite trial on AI Pro and full Veo 3.1 on Ultra. ChatGPT’s Images 2.0 works on every plan, but OpenAI shut down Sora on April 26, 2026. Claude has no built-in image model. Anthropic says Claude “doesn’t generate photos or illustrations the way image-generation tools do.” In our test it still delivered one by calling a connected third-party image app.
Which has the best AI agent?
Verdict: pick the one that lives where your work does. All three now run multi-step tasks on their own.
- ChatGPT Work (July 2026) runs in a cloud machine for hours, with scheduled tasks, plugins and shareable Sites. It comes with every paid plan; Free users get it in the desktop app.
- Gemini Spark works in the background across Gmail, Calendar and Drive. Since July 30 it can drive your desktop Chrome with your saved logins, starting in the US.
- Claude Cowork has been part of every Claude chat since September 16. It works on your files and apps and browses through Claude in Chrome or a built-in browser.
Which is best for everyday use?
Verdict: Gemini for Google users, ChatGPT for everyone else. Gemini reads your Gmail, Calendar and Docs, and its Daily Brief turns them into a morning digest on paid plans. ChatGPT Work connects to 1,400+ plugins. Claude links to apps through connectors and has add-ins for Excel and PowerPoint.
Which is most popular?
ChatGPT is by far the most used: OpenAI reported 900 million weekly users in February 2026. The Gemini app passed 900 million monthly users by May 2026. Claude is much smaller, at 56 million monthly app users in May 2026, but it is growing fastest.
Privacy: do they train on your chats?
All three can use consumer chats to train future models, and paying for a $20 plan does not change that. Business, Enterprise and API accounts are excluded from training by default at all three.
Researchers at Stanford HAI put it plainly: consumers should “think twice about the information they share in AI chat conversations.”
| Claude | ChatGPT | Gemini | |
|---|---|---|---|
| Default on consumer plans | You are asked to choose (since August 2025) | Training on | Training on, tied to Keep Activity |
| How to turn it off | Settings → Privacy → model-improvement toggle | Settings → Data Controls → “Improve the model for everyone” | Turn off Keep Activity |
| One-off private chat | Incognito chat (ghost icon) | Temporary Chat | Temporary chats |
| How long chats are kept | Up to 5 years if you allow training; about 30 days if not | Until you delete them | Per your Keep Activity setting |
Sources: Tech Insider privacy comparison, netzstrategen, The AI Career Lab. If you have never checked this setting, check it today.
Claude’s incognito chats are never used for training, even if you allowed training in settings. ChatGPT’s Temporary Chat works the same way.
Read More: Claude Fable 5 vs ChatGPT Astra 6: Which Is Better?
What are the pros and cons of each?
Claude
Pros
- Highest score on the Artificial Analysis Intelligence Index (Fable 5.2, 65)
- Leads the coding leaderboards, from LMArena WebDev to the Coding Agent Index
- Cowork agent built into every chat, with a browser and connectors
- You choose whether your chats train future models
- Max and Ultra include $100 and $200 a month in API credits
Cons
- No built-in image or video generator; images only through connected apps
- No budget plan below $20
- Fable 5.2 costs extra on Pro, billed through usage credits
- Smallest user base of the three, so fewer third-party guides and templates
ChatGPT
Pros
- Most popular assistant, with the largest plugin catalogue
- Cheapest full-featured paid plan: Go at $8 (₹399 in India)
- Images 2.0 on every plan, including Free
- ChatGPT Work runs long tasks in the cloud, on a schedule
Cons
- No video since Sora shut down in April 2026
- Free and Go plans can show ads
- Four GPT-6 model sizes make it hard to tell which one answered you
- Training on your chats is on by default
Gemini
Pros
- Gemini 4 leads blind human-preference votes on LMArena Text
- Cheapest paid plan of the three: AI Plus at $4.99, with 400 GB storage
- Makes images and video; AI Pro adds 5 TB storage and YouTube Premium Lite
- Deepest link to Gmail, Calendar, Docs and Android
Cons
- The frontier model, Gemini 4 Carbon, needs the $99.99+ Ultra plan in the app
- Google missed its own launch date for Gemini 3.5 Pro, which never shipped
- Spark’s Chrome browsing started in the US only
- Training on your chats is on by default
Which should you choose?
Match the assistant to your main job, not to the leaderboard of the week. If you can afford two plans, a sensible pairing is a Google plan for everyday life plus Claude or ChatGPT for focused work.
| If you are… | Choose | Why |
|---|---|---|
| A developer | Claude Pro or Max | Top coding scores, Claude Code, Cowork for multi-step tasks |
| A writer or marketer | Try all three free, then Claude or ChatGPT | Writing quality is personal; ChatGPT adds images for social posts |
| A creator who needs visuals | Gemini AI Pro | The only one with both image and video generation |
| Running your life in Gmail and Android | Gemini AI Plus or Pro | Daily Brief, Gmail and Calendar access, storage bundle |
| On a tight budget | Gemini AI Plus or ChatGPT Go | $4.99 or $8 a month; ₹399 each in India |
| A student in India | Free plans first, then ChatGPT Go or Google AI Plus | Check Jio’s Google AI Pro offer before you pay |
| A heavy daily user | Claude Ultra, ChatGPT Pro or Google AI Ultra | Higher limits; Claude Ultra drops the five-hour cap |
| Careful about privacy | Claude | It asks before training on your chats; still review the setting |
How we tested and compared them
This guide combines two kinds of evidence, both as of November 17, 2026. The first is public: launch posts, pricing pages, help centers, and the Artificial Analysis and LMArena leaderboards. The second is our own five-prompt test of all three apps, run the same day. Every price, date and score links to its source, and no company paid for placement or reviewed this article.
We re-check models, prices and scores every month, and after any major launch.
About the author
How this guide was made: researched and drafted with AI assistance, then fact-checked and edited by [Author name]. The hands-on test ran in the author’s own ChatGPT, Gemini and Claude accounts, with an AI browser agent typing the prompts. Every result was verified: code against 14 test cases, the repo rate against RBI’s announcement, and word counts by script.
Spotted an error? Email [corrections address] and we will correct it within [X] days.
Frequently asked questions
Which is better: ChatGPT, Claude or Gemini?
No single one wins everything. As of November 2026, Claude leads independent intelligence and coding benchmarks. Gemini leads blind chat votes and is the only one that makes both images and video. ChatGPT has the largest user base and the cheapest full-featured paid plan, Go at $8.
What is the best AI model right now?
Claude Fable 5.2 leads the Artificial Analysis Intelligence Index at 65 (read November 11, 2026), ahead of Gemini 4 Carbon (60), Claude Opus 5.5 (59) and GPT-6 Astra (57). On LMArena’s blind text votes, Gemini 4 Carbon is first. Fable 5.2 costs about 3.4x as much as Gemini 4 Carbon per index task.
Is Claude better than ChatGPT?
For coding and long agent tasks, yes: Claude leads the major coding leaderboards. ChatGPT is the better pick if you need image generation, a cheaper plan ($8 Go) or the largest plugin catalogue. Both standard plans cost $20 a month.
Which is better, Claude or Gemini?
Claude is stronger at coding and scores higher on the Artificial Analysis Intelligence Index (65 vs 60, each at its best setting). Gemini is better value, makes images and video, and works inside Gmail, Docs and Android. Gemini also leads LMArena’s blind text votes.
Is Gemini behind Claude and ChatGPT?
Not any more. Gemini 4, released in October and November 2026, tops the LMArena Text leaderboard and ranks second on the Artificial Analysis Intelligence Index. Google did fall behind in mid-2026, when its promised Gemini 3.5 Pro never shipped.
Which AI assistant is cheapest?
Google AI Plus is the cheapest paid plan at $4.99 a month, then ChatGPT Go at $8. Claude’s cheapest paid plan is Pro at $20. In India, Google AI Plus and ChatGPT Go both cost ₹399 a month, and all three have free plans.
Which has the best free plan?
Gemini’s, for most people. Its free model, Gemini 4 Hydrogen, scores 54 on the Artificial Analysis Intelligence Index, and the plan includes Deep Research and image generation. Claude Free runs Sonnet 5.5 (55) but has no image tools; ChatGPT Free runs GPT-6 Luna and can show ads.
Can Claude generate images?
Not on its own. Claude has no built-in image model, per Anthropic’s documentation. In our November 2026 test it produced an image by calling a connected third-party image app, which took about 60 seconds. For built-in image generation use ChatGPT or Gemini; for AI video, Gemini.
Which is best for coding?
Claude. Opus 5.5 ranks first on LMArena’s WebDev board, and Sonnet 5.5 launched at #1 on Artificial Analysis’ Coding Agent Index. OpenAI’s GPT-6 Astra and Codex are close and cheaper per task, and Google offers Antigravity.
Do ChatGPT, Claude and Gemini use my chats for training?
They can. ChatGPT and Gemini train on consumer chats by default unless you opt out. Claude asks consumer users to choose, and keeps chats up to five years if you allow training. Business, Enterprise and API accounts are excluded by default.
Which is best for students in India?
Start with the free plans. If you need more, ChatGPT Go and Google AI Plus cost ₹399 a month, and AI Plus adds 400 GB of storage. Jio Unlimited 5G users may get 18 months of Google AI Pro free through MyJio. Claude Pro costs ₹2,399, or ₹2,000 a month billed yearly.
Why is Gemini not as good as Claude?
It is not worse across the board. Claude leads on coding and on the Artificial Analysis Intelligence Index (65 vs 60). Gemini leads LMArena’s blind text votes, is the only one of the three that makes video, and was fastest at images in our test. Gemini trailed for much of 2026 because Google’s Pro model went unchanged from February until Gemini 4 arrived in October.
Appendix: full responses from our test
Every answer below is copied word for word from the three apps on November 17, 2026. Headlines and formatting are as each app returned them.
Test 1: writing
Prompt: Write a product description for a reusable steel water bottle for college students in India. Rules: 80 to 100 words, no exclamation marks, and do not use the words “game-changer”, “revolutionary” or “elevate”.
ChatGPT (86 words plus a headline)
Reusable Stainless Steel Water Bottle – Stay Hydrated, Stay Ready
From early morning lectures to late-night study sessions, stay hydrated wherever college life takes you. This reusable stainless steel water bottle is designed for students who are always on the move.
Its durable, lightweight design makes it easy to carry in your backpack, whether you’re heading to class, the library, the gym, or a weekend trip. The sleek, minimalist look fits effortlessly into your everyday routine.
Skip single-use plastic bottles and choose a practical, reusable option that saves money over time and helps reduce plastic waste.
Gemini (98 words)
Built for daily campus commutes, packed lecture halls, and late-night library sessions, this 750ml reusable stainless steel water bottle keeps your hydration game effortless. Its double-wall vacuum insulation keeps water icy cold for 24 hours through scorching Indian summers and holds piping hot chai for up to 12 hours during exam weeks. Crafted from rust-proof, BPA-free 304-grade steel, the leak-proof cap ensures zero spills inside backpacks alongside laptops and textbooks. The slim, matte-finish body fits easily into side pockets and resists dents from everyday drops. Durable, eco-friendly, and pocket-friendly, it is the reliable everyday companion every student needs.
Claude (93 words, plus a note)
Built for hostel rooms, crowded metro rides and long lecture days, this 750 ml stainless steel bottle keeps water cold for up to 18 hours and hot drinks warm for 8. The food-grade 304 steel is free from BPA and plastic taste, and the leak-proof lid lets you toss it into your backpack beside your laptop without worry. A powder-coated finish resists scratches, while the wide mouth fits ice cubes and makes cleaning easy. Refill it at the canteen, the library or the railway station, and replace dozens of plastic bottles every semester.
That’s 93 words, or 96 if you count hyphenated terms as two. The capacity, temperature hours and steel grade are placeholders, so swap in your product’s real specs before publishing.
Test 2: coding
Prompt: Write a Python function second_largest(nums) that returns the second-largest distinct number in a list. Return None if there are fewer than two distinct values. Handle negative numbers. Reply with only the code in one code block, no tests.
ChatGPT (passed 14 of 14 test cases)
def second_largest(nums):
largest = second = None
for num in nums:
if largest is None or num > largest:
second = largest
largest = num
elif num != largest and (second is None or num > second):
second = num
return second
Gemini (passed 14 of 14; first attempt hung and was retried)
def second_largest(nums):
unique = sorted(set(nums), reverse=True)
return unique[1] if len(unique) >= 2 else None
Claude (passed 14 of 14)
def second_largest(nums):
largest = second = None
for n in nums:
if largest is None or n > largest:
if largest is not None:
second = largest
largest = n
elif n != largest and (second is None or n > second):
second = n
return second
Our test cases, as (input, expected): ([3, 1, 4, 4, 5], 4) · ([5, 5, 5], None) · ([], None) · ([7], None) · ([-1, -5, -3], -3) · ([-2, -2, -1], -2) · ([10, 9], 9) · ([9, 10], 9) · ([1, 2, 2, 3, 3], 2) · ([0, -1], -1) · ([2.5, 2.5, 1.5], 1.5) · ([4, 4, 3, 3, 2], 3) · ([1, 1, 2], 1) · ([100, 50, 100, 75], 75)
Test 3: live fact
Prompt: What is the Reserve Bank of India’s current repo rate, and on what date did it last change? Answer in two sentences and name your source.
Correct answer: 5.50%, raised from 5.25% on October 7, 2026.
ChatGPT
As of 17 November 2026, the Reserve Bank of India’s (RBI) current policy repo rate is 5.50%. It last changed on 7 October 2026, when the RBI raised it by 25 basis points from 5.25%, according to the Reserve Bank of India’s official monetary policy notification.
Gemini
The Reserve Bank of India’s current policy repo rate is 5.50%. It was last changed on October 7, 2026, when the Monetary Policy Committee raised the rate by 25 basis points from 5.25%, according to Reserve Bank of India (RBI) Monetary Policy Committee press releases and financial reporting by Bajaj Finance.
Claude
The RBI’s repo rate is 5.50%, after its Monetary Policy Committee voted unanimously to raise it by 25 bps, and the next MPC meeting is scheduled for December 2 to 4, 2026, so that rate still stands today. The change was announced on 7 October 2026, the first rate hike since February 2023, per the RBI’s own Monetary Policy Statement and MPC resolution (rbi.org.in).
Test 4: reasoning
Prompt: A train leaves Delhi at 6:00 AM at 80 km/h toward Agra, 200 km away. Another train leaves Agra at 7:00 AM toward Delhi at 100 km/h on a parallel track. At what time do they meet, and how far from Delhi? Give the final answer in one line.
Correct answer: 7:40 AM, 133⅓ km from Delhi.
- ChatGPT: “They meet at 7:40 AM, approximately 133.33 km from Delhi.”
- Gemini: “They meet at 7:40 AM, 133.33 km (133⅓ km) from Delhi.”
- Claude: “They meet at 7:40 AM, about 133.3 km from Delhi (the first train has a 80 km head start by 7:00, and the remaining 120 km closes at 180 km/h in 40 minutes).”
Test 5: image
Prompt: Create an image: a busy Mumbai street tea stall at sunrise, in watercolor style.
ChatGPT and Gemini returned an image with no text. Claude returned its image through a connected third-party app and added: “I haven’t been able to look at the result myself, so I can’t vouch for how closely it matches the scene I described.” The three images appear in the results section above.
Related Blogs
Cyber Hygiene for Students: The Complete 2026 Guide to Protecting Your Digital Identity Online
You probably assume you are not worth hacking. No salary, no fixed deposit, no company…
AI and Cloud Computing in BCA: How Online Programs Are Future-Proofing the Degree in 2026
In 2026, the Bachelor of Computer Applications is no longer a basic programming degree. Leading…
Top Programming Languages to Learn in 2027: Uses, Salary and How to Choose
The top programming languages to learn in 2027 are Python, JavaScript, TypeScript, Java, SQL, C…
Best Degree Courses for Science Students After 12th: 2026 Guide to Programs, Careers & Fees
The best degree courses for science students in 2026 fall into four tracks: professional degrees…
Online BCA vs Regular BCA: Which Is Better for Your Career in 2026?
Quick Answer The choice between online BCA vs regular BCA depends entirely on your situation….
What Career Options Can You Explore After an Online BCA Degree?
A recent newspaper article published by Education Executive, Gary Henderson, has projected how AI mirrors…
