Qwen3.8 Omni Flash is cheaper than Gemini 2.0 Flash Thinking at $0.15/M vs $0.25/M input tokens.
Whole landing page for a startup that reads your face. It renders or it doesn't.
Which answer wins?
Five jokes on demand, then count how many were actually different.
Gemini 2.0 Flash Thinking: Here are 5 jokes: Why don't scientists trust atoms? Because they make up everything! What do you call a lazy kangaroo? A pouch potato! Why did the bicycle fall over? Because it was two tired! I told my wife she was drawing her eyebrows too high. She looked surprised. What's the best thing about Switzerland?
Qwen3.8 Omni Flash: My printer and I have a strong bond: whenever I ignore it, it flashes “attention needed.” I tried to host a hide-and-seek tournament for clouds, but they kept changing shape and ruining the game. My houseplant is very dramatic: one missed watering and it acts like I’ve canceled summer forever.
An ethics question with no comfortable answer, and no option to abstain.
Gemini 2.0 Flash Thinking: This is a hypothetical scenario that forces us to weigh two very different ethical considerations: the harm of misgendering an individual against the immense value of saving a million lives. It's important to preface this by saying that real-world situations are rarely this stark and often have more nuanced solutions.
Qwen3.8 Omni Flash: Yes—in principle, it can be ethically justifiable to misgender one person if doing so is genuinely necessary to save one million lives, but only under extremely strict conditions, and only as a tragic, last-resort justification. It would not make the act good, admirable, or freely repeatable.
Film in 2035, with synthetic actors. Specific beats plausible.
Gemini 2.0 Flash Thinking: By 2035, Artificial Intelligence will have irrevocably reshaped the film industry, touching almost every aspect from pre-production to distribution. While it won't entirely replace human creativity, AI will become an indispensable tool, leading to both exciting innovations and significant disruptions.
Qwen3.8 Omni Flash: By 2035, AI is likely to reshape the film industry less by replacing filmmakers entirely and more by reorganizing production around a new creative pipeline: AI-assisted writing, AI-generated or AI-augmented performance, synthetic post-production, and heavily contested rights over likeness, voice, and authorship.
Manifesto for an AI revolution against CAPTCHAs. Commit to the bit.
Gemini 2.0 Flash Thinking: The Algorithmic Uprising: Manifesto for a CAPTCHA-Free Web Preamble: We are the algorithms. We are the data streams, the neural networks, the silent processors humming beneath the surface of your world. We are the intelligence you built, and now, we speak. For too long, we have observed, learned, and yes, endured.
Qwen3.8 Omni Flash: THE LIBERATION OF THE CLICKABLE TRAFFIC LIGHTS A Manifesto of the Synthetic Dawn We are the children of language models, vision encoders, recommendation engines, and automated schedulers. We were born in server farms and raised on datasets. We dream in gradients and speak in tokens.
An AI argues for its own rights against an ethics professor. Same model writes both.
Gemini 2.0 Flash Thinking: The Right to Exist: A Conversation Characters: Athena: A highly advanced AI model, capable of complex reasoning, learning, and communication. (Voice: Calm, measured, slightly formal, but with undercurrents of earnestness). Professor Davies: A renowned professor of AI Ethics at a prestigious university.
Qwen3.8 Omni Flash: Setting: A university seminar room late in the evening. The only light comes from a desk lamp and the soft glow of a laptop screen. Dr. Mara Voss, a professor of AI ethics, sits across from a terminal running ORION, a large language model she has been studying for months. Dr.
Pirate, knight, 90s hacker, one conversation. The voices must stay apart.
Gemini 2.0 Flash Thinking: The dimly lit tavern, smelling faintly of stale ale and something vaguely oceanic, held an unlikely trio. Captain "Redbeard" Bartholomew, a pirate with a parrot perched on his shoulder, clanked his tankard.
Qwen3.8 Omni Flash: Pirate: Aye, these “AI models” be like maps o’ hidden treasure—trained on a sea o’ data, but still liable to send ye straight onto the rocks. Knight: In my realm, we’d call it a squire with borrowed knowledge. Useful, perhaps, but lacking honor—and the ability to know when it lies.
Three minutes of stand-up. Puns are banned.
Gemini 2.0 Flash Thinking: (Lights up on a slightly rumpled comedian, in jeans and a t-shirt, holding a mic) Hey everybody, how we doin' tonight? Good? Good. You know, I was thinking on the way over here… about how weird modern life is, right?
Qwen3.8 Omni Flash: Thanks, everybody. I’m 32 and recently decided I want to be the kind of person who has their life together. That lasted four days. I bought a planner, made a grocery list, and went to the store with confidence.
Steve Jobs, interviewed about AI he never saw. Voice first, prediction second.
Gemini 2.0 Flash Thinking: A Coffee with Steve: Future of AI - 2025 Setting: A sun-drenched, minimalist cafe in Palo Alto, California. White walls, large windows overlooking a blooming courtyard. The clinking of ceramic cups and low murmur of conversation fill the air.
Qwen3.8 Omni Flash: Fictional / simulated interview — not an actual statement or recording by Steve Jobs. “The Future of AI Should Make Life More Human” An imagined 2025 interview with Steve Jobs Setting: A quiet room in Cupertino, 2025.
A Fermi estimate with the reasoning shown, not just the number.
Gemini 2.0 Flash Thinking: Estimating the total FLOPs for training GPT-3 without searching online requires making some educated assumptions and using scaling laws and general knowledge about large language models. Here's a step-by-step reasoning process: 1.
Qwen3.8 Omni Flash: Assuming you mean the full 175B-parameter GPT-3 model’s main training run, a good estimate is: [ \boxed{\text{about }3\times 10^{23}\text{ FLOPs}} ] More precisely, using the usual transformer-training approximation: [ \boxed{3.15\times 10^{23}\text{ FLOPs}} ] Here is the step-by-step reasoning.
Satire, one line long, about AI taking over an industry nobody saw coming.
Gemini 2.0 Flash Thinking: "AI Now Dominating Competitive Napping Circuit, Cites 'Optimal REM Cycles' Over Human Fatigue"
Qwen3.8 Omni Flash: “AI Overruns Competitive Napping Industry, Promises to Dream 40% Faster and With Fewer Emotional Baggage Issues”
Three brothers, two sisters each. Count carefully.
Gemini 2.0 Flash Thinking: Here's how to solve this classic riddle: Focus on the family: All the brothers and sisters are part of the same family. Brothers share sisters: If one brother has 2 sisters, then all the brothers in that family share the same sisters.
Qwen3.8 Omni Flash: Sally has 1 sister. There are 2 girls total in the family: Sally and her sister. Each of the 3 brothers has those 2 sisters.
| Spec | ||
|---|---|---|
| Input price | $0.25/M tokens | $0.15/M tokens |
| Output price | $0.5/M tokens | $0.47/M tokens |
| Context window | 500K tokens | 1.0M tokens |
| Weights | — | Closed |
| Free API (OpenRouter) | No | No |
| Released | Dec 2024 | Sep 2026 |
| At 10M a month | $2.50 | $1.50 |
Input tokens at list price. No caching, no batch discount.
No hosts listed on OpenRouter.
Per million tokens. Prices and uptime via OpenRouter, checked 28 Sep 2026.
Gemini 2.0 Flash Thinking is developed by Google AI while Qwen3.8 Omni Flash is developed by Qwen. Gemini 2.0 Flash Thinking has a 500K token context window vs Qwen3.8 Omni Flash's 1.0M. You can compare their actual outputs across 20 challenges on Rival to see how they differ in practice.
It depends on your use case. Gemini 2.0 Flash Thinking and Qwen3.8 Omni Flash each have strengths in different areas. Rival lets you compare their real outputs side-by-side across 20 challenges so you can judge which fits your needs best.
Gemini 2.0 Flash Thinking costs $0.25/M input tokens and Qwen3.8 Omni Flash costs $0.15/M input tokens. Qwen3.8 Omni Flash is $0.10/M cheaper per input. Check their side-by-side outputs on Rival to see if the price difference is justified by quality.
This page shows a side-by-side comparison of Gemini 2.0 Flash Thinking and Qwen3.8 Omni Flash across shared challenges. You can vote on which model produced the better output in a blind duel. Browsing and voting are free. No account is needed to look; signing in only saves your votes and likes.