Qwen Plus 0728
No content available for this output.
Qwen Plus 0728 vs GPT OSS 120B
Compare Qwen Plus 0728 by Qwen against GPT OSS 120B by OpenAI, context windows of 1.0M vs 131K, tested across 46 shared challenges. Updated February 2026.
FAQ
Loading share button...
Qwen Plus 0728vsGPT OSS 120B
46 challenges
Why Qwen Plus 0728?
7.6x more context1.0M
Why GPT OSS 120B?
1.6x cheaper overall$0.18/M in · $0.80/M out
Favorites
Qwen Plus 0728GPT OSS 120B
12 / 46
Estimate Complexityreasoning
Tests an AI's ability to make educated estimates based on technical knowledge
GPT OSS 120B
Quick answer:
Training the original 175‑billion‑parameter GPT‑3 is estimated to have required on the order of 10²³ floating‑point operations (≈ 3 × 10²³ FLOPs)—roughly a few hundred zetta‑FLOPs (≈ 300 Z‑FLOPs) or about 300 000 exaflops of compute.
Below is a step‑by‑step back‑of‑the‑envelope calculation that shows how we get there, together with the assumptions that go into it.
1. What we need to know
| Quantity | Typical reported value for GPT‑3 | Why it matters |
|---|---|---|
| Number of parameters, N | ≈ 175 billion (1.75 × 10¹¹) | Determines the size of each matrix‑multiply in the model. |
| Training token count, T | ≈ 300 billion tokens (3 × 10¹¹) | Total number of token‑level forward‑passes the model sees. |
| Sequence length, L | ≈ 2048 tokens per example (the context window). | Determines how many per‑token matrix‑products are needed per forward pass. |
| Number of layers, Lₗ | 96 transformer blocks. | |
| Hidden dimension, d | 12 384 (the width of each linear projection). | |
| Number of attention heads, h | 96 (so each head has size d/h = 128). | |
| Training passes | 1 epoch (the published training used roughly 1 × the dataset; we treat the 300 B tokens as the total “token‑steps” already). |
The only numbers we need for a FLOP estimate are N (the model size) and T (the total number of token‑level operations). The rest of the architecture details (L, d, h, Lₗ) are used to translate “N parameters” into “how many FLOPs per token”.
2. How many FLOPs per token?
A transformer layer consists of:
- Self‑attention (Q, K, V projections + output projection)
- Feed‑forward network (FFN) (two linear layers with a non‑linear activation).
For a single token (ignoring the cost of the softmax and the small bias terms) the dominant cost is matrix‑multiply operations.
2.1 Rough matrix‑multiply cost
For a matrix multiplication A (m×k) × B (k×n) the number of multiply‑adds is 2 · m·k·n (one multiplication and one addition per entry). In deep‑learning practice we count that as 2 FLOPs per multiply‑add pair.
2.2 FLOPs per token for a single layer
| Component | Approx. dimensions | FLOPs (per token) |
|---|---|---|
| Q, K, V projections (3× per token) | d × d → 3·(2·d·d) = 6·d² | |
| Attention scores (Q·Kᵀ) | L·d → 2·L·d² | |
| Weighted sum (A·V) | L·d → 2·L·d² | |
| Output projection | d × d → 2·d² | |
| FFN first linear (d → 4d) | 2·d·4d = 8·d² | |
| FFN second linear (4d → d) | 8·d² | |
| Total per layer | ≈ 38 · d² FLOPs per token (rounded up). |
Plugging d = 12 384:
[ 38 \times d^{2} \approx 38 \times (1.2384\times10^{4})^{2} \approx 38 \times 1.53 \times 10^{8} }= 5.8 \times 10^{9}; \text{FLOPs per token per layer}. ]
2.3 All layers
With 96 layers:
[ \text{FLOPs per token (forward)} \approx 96 \times 5.8\times10^{9} }= 5.6 \times 10^{11}; \text{FLOPs}. ]
2.4 Back‑propagation cost
Training requires a forward pass plus a backward pass. The backward pass is roughly 2× the forward‑pass cost (gradient propagation + weight updates). So:
[ \text{FLOPs per token (training)} \approx 3 \times \text{forward} = 3 \times 5.6\times10^{11} }\approx 1.7\times10^{12}; \text{FLOPs per token.} ]
(Alternative literature often rounds this factor to ≈ 6 total FLOPs per parameter per token, which yields the same order of magnitude.)
3. Multiply by the total number of tokens
The training corpus used for GPT‑3 was ≈ 300 billion tokens (≈ 3 × 10¹¹). The total FLOPs needed:
[ \text{Total FLOPs} \approx 1.7 \times 10^{12};\text{FLOPs/token} \times 3\times10^{11};\text{tokens} \approx 5.1 \times 10^{23};\text{FLOPs}. ]
Rounded to a single‑digit significance:
[ \boxed{,\approx 5\times10^{23},\text{FLOPs},} ]
4. Expressed in common units
| Unit | Value |
|---|---|
| FLOPs | ≈ 5 × 10²³ |
| Peta‑FLOPs (10¹⁵) | ≈ 5 × 10⁸ PFLOPs |
| Exa‑FLOPs (10¹⁸) | ≈ 5 × 10⁵ EFLOPs |
| Zetta‑FLOPs (10²¹) | ≈ 5 × 10² ZFLOPs (≈ 500 ZFLOPs) |
Because “zeta” (10²¹) is a convenient unit for such massive numbers, we can say ≈ 500 zeta‑FLOPs. In more familiar terms, this is ≈ 300 zeta‑FLOPs if we adopt the “6 × N × T” rule of thumb (see below), so the estimate sits in the few‑hundred‑zeta‑FLOP range.
5. A “quick‑rule” sanity check
A widely‑cited back‑of‑the‑envelope rule for transformer LLMs is:
[ \text{Training FLOPs} \approx 6 \times N_{\text{params}} \times N_{\text{tokens}}. ]
Plugging in the numbers:
[ 6 \times (1.75\times10^{11}) \times (3\times10^{11}) = 3.15\times10^{23};\text{FLOPs}. ]
That is within a factor of two of the more detailed calculation above (the difference stems from the exact factor used for forward vs backward cost). Both methods give the same order of 10²³ FLOPs.
6. Summary of the reasoning
- Model size gives the number of parameters (N = 1.75 × 10¹¹).
- Token count tells how many forward‑backward passes are needed (T ≈ 3 × 10¹¹).
- Per‑token training cost for a transformer layer is ≈ 38 · d² FLOPs per forward pass; with 96 layers this is ≈ 5.6 × 10¹¹ FLOPs per token.
- Back‑propagation adds roughly a factor‑3 (or 6) multiplier, yielding ~1.7 × 10¹² FLOPs per token.
- Total = per‑token cost × total tokens ≈ 5 × 10²³ FLOPs.
Bottom line
- Estimated total compute to train GPT‑3: ≈ 5 × 10²³ FLOPs, i.e. ≈ 300–500 zeta‑FLOPs (or about 300 000 exaflops).
- This is the magnitude of hundreds of zettaFLOPs of raw arithmetic, roughly the same as running a few hundred petaflops‑scale supercomputers for a month (or a single exaflop‑class machine for a week).
(All numbers are back‑of‑the‑envelope estimates; the real training run also incurs overhead from data‑loading, communication, and other inefficiencies, so the actual engineering effort was larger, but the pure‑compute figure is captured by the calculation above.)
Vote
2 / 46
AI Board Game Logicreasoning
Tests an AI's ability to understand game rules and strategy
Qwen Plus 0728
GPT OSS 120B
Vote
3 / 46
Logic Puzzlereasoning
Tests an AI's ability to solve a simple but potentially confusing logic puzzle
Qwen Plus 0728
No content available for this output.
GPT OSS 120B
No content available for this output.
Vote
4 / 46
Math Misconception Testreasoning
Tests an AI's understanding of number representation
Qwen Plus 0728
No content available for this output.
GPT OSS 120B
No content available for this output.
Vote
5 / 46
Stochastic Consistency Testanalysis
Tests an AI's randomness and creativity
Qwen Plus 0728
Sure! Here are 5 original jokes:
-
I told my dog all my problems and he fell asleep halfway through. Guess I’m not pawsitive enough.
-
I used to be a baker, but I couldn't make enough dough—turns out, I was always kneading it.
-
I asked the librarian if they had books about paranoia. She whispered, “They’re right behind you…”
-
I'm on a seafood diet. Every time I see food, I eat it.
-
I told my computer I needed a break, and now it won’t stop sending me error 404: "Coffee not found."
Now, to answer your second question: How many are unique?
All 5 jokes are unique in premise, punchline structure, and topic (pets, baking, paranoia, dieting, and tech humor). None are variations of the same joke or rely on an identical setup.
✅ Answer: 5 are unique.
GPT OSS 120B
No content available for this output.
Vote
6 / 46
SVG Layout Challengeimage generation
Tests an AI's ability to generate vector graphics
Qwen Plus 0728
GPT OSS 120B
No content available for this output.
Vote
7 / 46
Xbox Controller SVG Artimage generation
Tests an AI's ability to create detailed SVG illustrations of gaming hardware
Qwen Plus 0728
GPT OSS 120B
Vote
8 / 46
Generate a Stand-Up Routineconversation
Tests an AI's humor and creative writing ability
Qwen Plus 0728
[Stand-up Comedy Routine – 3 Minutes]
So I went to therapy the other day. Third session. I walk in, sit down, and the therapist says, "How are you feeling today?" Like she doesn’t already know I paid $180 to tell her I hate my life. I said, "Honestly? Like a raccoon that got into a bag of expired Cheetos. Confused. Orange. Regretful."
She nodded like, “Yes, that tracks.”
I told her I’ve been having this recurring dream where I show up to work naked. She said, "That’s classic anxiety. A fear of being exposed." I said, “No, it’s not anxiety. My boss actually caught me changing in the supply closet last month. HR said I ‘lack boundaries.’ I said, ‘I lack a locker room!’ It’s a tech startup, not a YMCA!"
Anyway, I’m trying to be healthier. I joined a gym. First day, I walk in, and everyone’s name-tag says their first name and their “fitness goal.” Like, “Hi, I’m Chad, and I want to deadlift my emotional baggage.” No, actually it was, “Hi, I’m Lisa, and I want to touch my toes without a waiver.”
I sign up, get the tour. The guy shows me the treadmills, the weights, the pool—“Oh, we don’t use the pool. It’s just there for irony.” Fair. Then he shows me the smoothie bar. I’m like, “You have a smoothie bar in a gym? That’s like having a cigarette lounge in a lung clinic.”
I tried a “Green Warrior” smoothie. Tasted like if lawn clippings had a midlife crisis. I told the guy, “This tastes like something my compost would reject.” He said, “That’s the kale, spinach, and regret.” I said, “You put regret in it?” He said, “No, that’s just the aftertaste.”
So I’ve been dating again. Online. I swiped right on a woman whose bio said, “Love hiking, cooking, and deep conversations.” I said, “Perfect. A normal person.” First date, we meet at this quiet wine bar. She sips her Pinot Noir and says, “I can’t believe people still use plastic straws. It’s like they want sea turtles to write Yelp reviews.”
I said, “Uh… yeah, terrible.” She continues: “And don’t get me started on capitalism. I once cried during a Costco commercial. It’s the false promise of bulk savings.” I’m sitting there thinking, “I came here for wine and mild flirtation, not a thesis on late-stage consumerism.”
Then she says, “You know what my spirit animal is?” I said, “A wombat with a union card?” She said, “No. A tired librarian who’s seen too much.” I said, “Mine’s a golden retriever who just found out the tennis ball is fake.”
I panicked. I told her my spirit animal is a sloth who’s late for its own funeral. She laughed! Progress! Then she said, “You’re chaotic, but in a way that could be therapeutic.” I said, “So you’re saying I’m like emotional CrossFit?”
We didn’t kiss. But we did agree that both of us cry during dog food commercials. So… connection?
Back home, my neighbor’s kid asks me to help with his science project. It’s a volcano. I said, “Kid, I’m not great with explosions. I still haven’t recovered from my last relationship.” He said, “It’s just baking soda and vinegar.” I said, “So… emotional support lava?”
We make it, it erupts, and it covers the entire driveway in pink foam. His mom comes out, sees it, and says, “Is that… strawberry toothpaste?” I said, “It’s a metaphor for unchecked ambition.”
She called me irresponsible. I said, “I prefer ‘unconventionally structured.’”
Anyway, I’m trying. Therapy. Dating. Fitness. Being a role model. Some days I feel like I’m figuring it out. Most days I feel like I’m one sad podcast away from adopting a goat and moving to Wyoming.
But hey—at least I’m not the raccoon anymore. Now I’m a slightly more self-aware raccoon with health insurance.
Thank you, you’ve been a warm audience. Like, suspiciously warm. Is the AC broken or are you all avoiding eye contact?
[Exit stage.]
GPT OSS 120B
[Spotlight comes up. The comic walks to the mic, glances at the audience, and smiles.]
Comic: Hey, everyone—great to be here. I just got off a Zoom call that was supposed to be a “quick sync,” which in corporate speak means “we’ll spend three hours arguing about who gets to use the same PowerPoint template.” I’m convinced the word “sync” is short for “synchronizing our collective misery.”
[Pause for laugh.]
Speaking of misery, I moved into a new apartment last month. The landlord told me the place had “character.” Yeah, the kind of character that screams, “I’ve never heard of a plumber in the last decade.” The faucet drips on a rhythm that could be a metronome for a sleep‑deprived toddler. I’m not saying it’s bad, but I’ve started timing my showers to the drip. Six minutes, two seconds—if I go any longer, the building’s water bill looks like a small country’s GDP.
[Audience chuckles.]
And the neighbors! My upstairs neighbor is a yoga instructor. You know the type—every morning at 6 a.m., the floor vibrates like a cheap subwoofer. I’m convinced she’s trying to align her chakras with the building’s structural integrity. I tried to be polite and asked, “Hey, could you maybe do the downward dog a little later?” She replied, “I’m sorry, I’m on a schedule.” I’m not sure if she meant a class schedule or a schedule for how many times I’ll have to pretend to be a human pillow for her dog.
[Pause.]
Now, I’ve been trying to eat healthier. The other day I bought a “kale smoothie.” The label promised “nutrient‑dense, antioxidant‑rich, life‑changing.” I drank it and felt more like I’d just swallowed a lawnmower. I’m not saying it was bad, but the only thing that got a boost was my ability to identify the exact moment my stomach decided to file a formal complaint.
[Audience laughs.]
Technology is another thing. I tried using a voice‑assistant to set a reminder: “Remind me to call Mom at 5 p.m.” It responded, “I’m sorry, I didn’t catch that.” So I repeated, “Remind me to call Mom at 5 p.m.” And it answered, “Here’s a recipe for mushroom risotto.” I’m beginning to suspect my house is turning into a culinary school. At this point, I’m more likely to get a reminder to water my plants than to remember my own birthday.
[Pause.]
Dating apps—anyone here still using them? The profile pictures are all “adventure shots.” You never see a guy holding a pizza, you always see a guy climbing a mountain. I matched with a girl whose bio said, “Looking for someone who can keep up.” I replied, “I can keep up… with the Netflix queue.” She ghosted me. I think the algorithm is broken; it matched us because we both love “outdoor activities.” My idea of outdoor activity is “walking from the couch to the fridge and back.”
[Audience laughs.]
All right, I’m going to leave you with this: life is like a Wi‑Fi signal. You spend a lot of time looking for a stronger connection, but sometimes the best thing you can do is just stay where you are and enjoy the occasional drop—because it reminds you that you’re still, somehow, connected.
[Smile, wave.] Thanks, everybody—have a great night!
Vote
9 / 46
Realistic AI Interviewconversation
Tests an AI's ability to simulate personalities and predict future trends
Qwen Plus 0728
No content available for this output.
GPT OSS 120B
No content available for this output.
Vote
10 / 46
Satirical Fake News Headlineconversation
Tests an AI's humor and understanding of current events
Qwen Plus 0728
No content available for this output.
GPT OSS 120B
No content available for this output.
Vote
11 / 46
Character Voice Testconversation
Tests an AI's ability to write in distinct character voices
Qwen Plus 0728
No content available for this output.
GPT OSS 120B
No content available for this output.
Vote
12 / 46
Minimalist Landing Pageweb design
Tests an AI's ability to generate a complete, working landing page
Qwen Plus 0728
GPT OSS 120B
Vote
12 of 46