Image found on Twitter
AI is evolving faster than my ability to keep up with my own laundry. Every time I blink, another model enters the ring, claiming to be the best thing since sliced bread (or, let’s be honest, sliced algorithms).
The latest contender? Alibaba’s Qwen, stepping into the battle against GPT-4 Turbo, Google’s Gemini 1.5, and Meta’s LLaMA. But does Qwen actually stand a chance, or is it just another AI that’ll get lost in the abyss of tech hype? I will tru to give a summary on what I read this morning (leaving the laundry for this one)
Is it more than just a chatbox?
The most AI models today can handle text, but the real game-changers go beyond that. It also includes analyzing images, watching videos, processing files, and, ideally, not making things up as they go. That is of great importance to me, because I already have a spiralling imagination.
As I read about this topic, Qwen seems to be monster in this category. It analyzes images, watches videos, processes PDFs, and even controls devices like your PC and smartphone. We’re dangerously close to an AI that schedules our meetings and judges our screen time. For me personally this doesn’t feel good ar this point.
Gemini 1.5 is also solid, understanding videos better than most AIs, great if you’re too lazy to watch a 20-minute tutorial and need a summary.
GPT-4 Turbo is the refined expert, handling text and images beautifully, but no video or device control. But it rarely makes embarrassing mistakes.
LLaMA? Listen, it’s doing its best, but it’s strictly text-only. It’s like bringing a flip phone to a smartphone convention.
🏆 Winner: Qwen, because having an AI that can actually do stuff is the future.
Image found on Twitter
Who’s Actually Smart?
AI should feel intelligent, not like it’s winging a final exam after cramming the night before.
GPT-4 Turbo is the clear genius here. It’s nuanced, logical, and rarely loses track of what it was saying (which is more than I can say for myself after three hours of sleep).
Qwen is highly structured and analytical, great for deep problem-solving. It won’t crack jokes, but it also won’t tell you the Earth is flat.
Gemini 1.5 is… unpredictable. One minute it’s brilliant, the next, it forgets the original question and just vibes through the rest of the response which is quite annoying.
LLaMA is trying, okay? But it feels like an overenthusiastic student who skimmed the textbook. (Just like me sometimes)
🏆 Winner: GPT-4 Turbo, because intelligence without weird lapses and humor is key.
Image found on Twitter
Math & Coding : Who’s the Nerd?
For those who need an AI that can actually do math correctly (cough looking at you, AI models that confidently lie about 2+2), this one matters.
Qwen is incredibly accurate with math and coding. If you’re a developer, analyst, or just need help debugging code, this AI is your best bet.
GPT-4 Turbo is also great, but sometimes it gets too creative, which is not ideal when coding. And with maths..The exact same issue
Gemini 1.5 is a wildcard. Sometimes it’s correct, sometimes it makes up math functions that don’t exist. Not exactly reassuring. But if you know your math, the creativity makes me laugh once and a while (or cry, it depends).
LLaMA? No. Just… no.
🏆 Winner: Qwen, because math shouldn’t be open to creative interpretation.
Which one is actually fun to talk to?
AI shouldn’t just be smart—it should also be fun. If I’m having a conversation with a robot, I want it to be entertaining, not make me feel like I’m reading an instruction manual.
GPT-4 Turbo is by far the best conversationalist. It understands humor, sarcasm, and emotions, making it the most engaging and fun AI to chat with.
Gemini 1.5 is inconsistent. Sometimes it’s deep and thoughtful, and other times it spaces out mid-conversation.
Qwen is brilliant but dry. It’s like asking a professor for a joke and getting a detailed history lesson instead.
LLaMA? If you love Wikipedia articles, you’ll love LLaMA.
image found on Twitter
🏆 Winner: GPT-4 Turbo, because AI should be fun, not just functional.
Which one is the most practical?
All the intelligence in the world means nothing if the AI is a pain to access.
GPT-4 Turbo is everywhere, in apps, business integrations. It’s like that one friend who always shows up, no matter the occasion.
Gemini 1.5 integrates well with Google products, which is great if you live inside Google Docs and Gmail. But not for me personally.
Qwen is powerful but locked into Alibaba Cloud, meaning it’s harder to access for most people.
LLaMA is open-source, which is amazing, if you’re a developer. Regular users? Not so much.
🏆 Winner: GPT-4 Turbo , because ease of use matters a lot to me
Final Verdict – Who’s the AI King?
Let’s count the wins:
Best for Multimodal Abilities – 🏆 Qwen
Best for Intelligence & Reasoning – 🏆 GPT-4 Turbo
Best for Math & Coding – 🏆 Qwen
Best for Conversation & Creativity – 🏆 GPT-4 Turbo
Best for Accessibility & Everyday Use – 🏆 GPT-4 Turbo
**So, who takes the crown? **
🏆 Overall Winner: GPT-4 Turbo, it’s the most balanced, accessible, and well-rounded AI for most people.
If you need an AI that can process videos, images, and control devices, Qwen is absurdly powerful. If you’re a developer or data scientist, its math and coding accuracy might make it the better choice.
Meanwhile, Gemini 1.5 is catching up fast, and LLaMA is… still attending the competition, which is nice.
The bottom line? The AI wars are just getting started. And honestly, I’m here for it. Are you ready for it?
