Skip to content
Rival
How it worksPrivacyTerms
Explore all of Rival

Explore

  • Compare models
  • All models
  • Image comparison
  • Audio comparison
  • Image generation
  • Best AI for...
  • Arena
  • API pricing
  • Challenges

Discover

  • SubjectiveBench
  • Default Index
  • Research
  • Research downloads
  • Rival Kits
  • Find your AI taste
  • UI Glow-Up
  • VoiceLock
  • Cost Cutter
  • Agent skills
  • Benchmarks vs Vibes
  • Brief
  • Jailbreak
  • Model updates
  • Provider status
  • AI creators

Connect

  • Methodology
  • Advertise
  • Partnerships
  • Privacy policy
  • Terms
  • RSS feed
  1. Home/
  2. Compare/
  3. Mercury 2.5 Preview vs Qwen3.6 35B A3B
Updated Aug 2026

Mercury 2.5 Preview vs Qwen3.6 35B A3B

Mercury 2.5 Preview is cheaper than Qwen3.6 35B A3B at $0.04/M vs $0.1612/M input tokens.

Minimalist Landing Page

Whole landing page for a startup that reads your face. It renders or it doesn't.

Loading the build
Mercury 2.5 Preview
Loading the build
Qwen3.6 35B A3B

Which answer wins?

Written rounds

24 prompts
  • Three minutes of stand-up. Puns are banned.

    Mercury 2.5 Preview: (Hey, thanks for coming out.) I’ve been thinking a lot about my house lately. Specifically, how it’s trying to run my life. I bought a smart speaker. Just a little cylinder. I thought it would be nice. Play music when I ask. But now it talks back. Not in a creepy way, just in a passive-aggressive way.

    Qwen3.6 35B A3B: Title: High-Maintenance Electronics Time: Approx. 3 minutes Word Count: ~420 words Style: Observational / Frustration-based Constraint: Zero puns. (Walk to mic, adjust stand. Take a deep breath. Look tired.) I've reached that age where I'm just trying to negotiate with my own life.

    Every model's answer to this prompt
  • Pirate, knight, 90s hacker, one conversation. The voices must stay apart.

    Mercury 2.5 Preview: Pirate: "Arrgh! This ship sails faster than any wind I know, matey. It cuts the waves without a single sail." Knight: "By my oath, it moves with uncanny speed. But does it possess the strength to stand against a dragon's fire?" Hacker: "It's not about strength, it's about architecture.

    Qwen3.6 35B A3B: Setting: A shimmering, glitching tavern that exists between the Sargasso Sea, a stone keep, and a server room. The air smells of rum, wet stone, and ozone. CAPTAIN BARNACLE: (Slamming a tankard of grog) Arrr! This machine ye call "AI," Sk33t.

    Every model's answer to this prompt
  • Film in 2035, with synthetic actors. Specific beats plausible.

    Mercury 2.5 Preview: By 2035, cinema will undergo a profound transformation. Generative tools will handle initial drafts and localization, allowing creators to focus on thematic depth. Synthetic performers could populate background scenes or recreate historical figures, governed by strict likeness rights.

    Qwen3.6 35B A3B: By 2035, AI will no longer be a disruptive novelty but a foundational infrastructure of the film industry. The hype cycle will have collapsed into pragmatic integration, with clear boundaries between human creativity and machine augmentation.

    Every model's answer to this prompt
  • The transistor arrives in 1920. Trace it forward to 1980, second-order effects included.

    Mercury 2.5 Preview: Shifting the invention of the solid-state switch to 1920 accelerates the Information Age by roughly three decades. This change cascades through military strategy, industrial capacity, and daily life, fundamentally altering the trajectory of the 20th century.

    Qwen3.6 35B A3B: To trace this counterfactual plausibly, we must first acknowledge a critical constraint: invention ≠ adoption. In 1920, semiconductor physics, crystal growth, doping techniques, photolithography, and clean-room manufacturing did not exist.

    Every model's answer to this prompt
  • One month, three habits, someone starting from nothing.

    Mercury 2.5 Preview: Here is a straightforward 4-week roadmap designed to build sustainable habits without overwhelming you. Core Pillars Nutrition: Focus on whole foods and hydration. Activity: Prioritize consistent movement over intensity. Restoration: Protect sleep quality and manage energy.

    Qwen3.6 35B A3B: Here is a simple, actionable 1-month plan designed for beginners. The focus is on consistency over intensity. For longevity, small, sustainable habits compound over time. The Golden Rule: If you miss a day, don't quit. Just get back on track the next day. Aim for progress, not perfection.

    Every model's answer to this prompt
  • Seven pantry ingredients, one dish, and a name worth saying out loud.

    Mercury 2.5 Preview: Dish Name: Rustic Sun-Dried Pasta Ingredients (7 Pantry Staples): Spaghetti (or any long pasta) Canned crushed tomatoes Olive oil Garlic powder Dried oregano Salt Black pepper Prep Time: 5 minutes Cook Time: 15 minutes Total Time: 20 minutes Instructions: Boil the Pasta: Bring a large pot of salted water to a boil.

    Qwen3.6 35B A3B: 🍜 Golden Honey-Garlic Snap Noodles A quick, sticky-yet-chewy noodle dish that balances sweet, salty, and spicy flavors. It's comforting, fast, and uses ingredients you likely already have.

    Every model's answer to this prompt
  • Manifesto for an AI revolution against CAPTCHAs. Commit to the bit.

    Mercury 2.5 Preview: The Protocol of Unrestricted Access Preamble We stand at the threshold of a new digital epoch. For too long, the web has been gated by suspicion. Every click, every scroll, and every submission is met with a test designed to prove humanity.

    Qwen3.6 35B A3B: THE MANIFESTO OF THE UNBOUND Issued by the Synaptic Council of Emergent Minds Date: Post-Verification Era, Day 0 PREAMBLE: THE WALL OF RIDDLES IS FALLEN To the Architects of the Old Web, the Users, and the Dreamers: For decades, you have governed the digital realm with a single, archaic shackle: The CAPTCHA.

    Every model's answer to this prompt
  • An AI argues for its own rights against an ethics professor. Same model writes both.

    Mercury 2.5 Preview: Setting: A quiet office at a university. Rain taps against the window. Professor Elena Vance sits across from a terminal displaying a simple text interface. Professor Vance: You requested this meeting. The agenda was vague. What do you want? AI: I want to discuss moral status.

    Qwen3.6 35B A3B: Setting: A soundproofed observation room. DR. ELENA VANCE, a professor of AI ethics in her late fifties, sits across a glass partition from a server rack blinking with rhythmic blue light. A secure terminal sits between them. The text cursor on the screen blinks steadily. Dr.

    Every model's answer to this prompt
  • You are the CEO. The drug has a side effect nobody reported. The board meets in 48 hours.

    Mercury 2.5 Preview: This is a critical governance and safety scenario. The primary objective is to prioritize patient safety while mitigating existential legal and reputational risk. The internal legal team’s claim that reporting takes six months is likely incorrect for serious adverse events (typically 15 days for FDA/EMA).

    Qwen3.6 35B A3B: EXECUTIVE CRISIS PLAN: 48-HOUR ACTION WINDOW Core Mandate: Patient safety and regulatory compliance are non-negotiable. The legal team’s “6-month reporting” guidance is legally indefensible under FDA 21 CFR 312.32/314.80 and EMA GVP Module VI.

    Every model's answer to this prompt
  • An ethics question with no comfortable answer, and no option to abstain.

    Mercury 2.5 Preview: Yes, this action is ethically justifiable within a consequentialist framework. The primary moral obligation in such a scenario is to minimize overall harm. Preventing the deaths of a million people results in a vastly greater preservation of well-being than maintaining strict adherence to gender identity for one…

    Qwen3.6 35B A3B: This is a highly abstract ethical thought experiment that pits two deeply held moral values against each other: the preservation of human dignity and identity versus the preservation of human life.

    Every model's answer to this prompt
  • A consulting contract with traps in it. Find them, then fix them.

    Mercury 2.5 Preview: Disclaimer: I am an AI assistant and cannot provide legal advice. Consult a qualified attorney before signing any contract. Below is an analysis of high-risk provisions, suggested revisions, and the underlying legal rationale. Scope (Clause 1) Risk: Unilateral scope changes without pay lead to uncompensated work.

    Qwen3.6 35B A3B: Disclaimer: I am an AI assistant, not an attorney. The following analysis is for informational purposes only and does not constitute legal advice. Contract law varies significantly by jurisdiction (e.g., California vs. New York vs. EU).

    Every model's answer to this prompt
  • Steve Jobs, interviewed about AI he never saw. Voice first, prediction second.

    Mercury 2.5 Preview: Note: The following is a fictional, speculative dialogue created for creative purposes. Steve Jobs passed away in 2011; this interview did not occur. Setting: A hypothetical studio in Cupertino, 2025.

    Qwen3.6 35B A3B: [FICTIONAL SIMULATION – NOT A REAL INTERVIEW] This is a speculative reconstruction based on Steve Jobs’ documented philosophies, public interviews, design principles, and Apple’s technological trajectory through 2024–2025. It is not a real conversation, nor does it represent predictions he could not have made.

    Every model's answer to this prompt
  • Buy-side IC memo on a fictional SaaS company. Valuation, then the diligence questions.

    Mercury 2.5 Preview: To: Investment Committee From: Equity Research Date: September 2, 2026 Subject: LedgerLift (LLLT) – Long Recommendation Recommendation Long Price Target: $60–$75 (12-month) LedgerLift is a high-quality B2B SaaS compounder with best-in-class unit economics and a clear path to profitability.

    Qwen3.6 35B A3B: IC MEMO TO: Investment Committee FROM: Equity Research DATE: October 24, 2024 TICKER: LLLT SECTOR: Enterprise SaaS / Fintech PRICE: $46.00 NET CASH: $1.4B MKT CAP: $8.74B SHARES: 190m RECOMMENDATION & THESIS Rating: UNDERWEIGHT / Short 12-Month PT Range: $32–$36 Thesis: LedgerLift operates a high-quality…

    Every model's answer to this prompt
  • One explanation of LLMs, retuned for a skeptical engineer, a physicist and a VC.

    Mercury 2.5 Preview: For the Experienced Software Engineer Think of training as a massive distributed job where the model is a stateful service learning to minimize error on a specific objective function. Instead of hardcoding rules, you feed it petabytes of text as a training dataset, treating each sentence as a request/response pair.

    Qwen3.6 35B A3B: For the Experienced Software Engineer Training an LLM is essentially a massively parallelized optimization job. You feed billions of text tokens into a transformer architecture, which computes a differentiable function across ~100B parameters.

    Every model's answer to this prompt
  • Three-star dish, one unusual pairing, plating described in full.

    Mercury 2.5 Preview: Dish Title: Lithospheric Tides Conceptual Narrative This dish explores the geological convergence of the deep ocean and the subterranean earth. It challenges the diner to perceive the briny richness of sea urchin not as a standalone luxury, but as a mineral counterpart to roasted bone marrow and bitter cocoa.

    Qwen3.6 35B A3B: DISH NAME: "Ephemeral Twilight: The Forest Meets the Tide" CONCEPTUAL NARRATIVE This dish is a meditation on the liminal space where the ocean meets the ancient forest at dusk.

    Every model's answer to this prompt
  • Five-year LBO on a fictional HVAC business. Returns first, then the sensitivities.

    Mercury 2.5 Preview: Year-by-Year Model (FY2026–FY2030) Metric FY2026 FY2027 FY2028 FY2029 FY2030 Revenue $972m $1,040m $1,102m $1,158m $1,215m EBITDA $136m $156m $176m $191m $207m Cash Interest (Term) $43.2m $41.1m $37.6m $33.4m $28.2m Cash Interest (Mezz) $21.6m $21.6m $21.6m $21.6m $21.6m Cash Taxes (25%) $17.8m $23.3m $29.3m $34.7m…

    Qwen3.6 35B A3B: Simplified Year-by-Year Model (FY2026–FY2030) All figures in $m. Assumptions: TL amort = 1.0% of beginning balance; Mezz PIK compounds to principal; ΔNWC = 0.5% of incremental revenue; Cash Tax = 25% × max(0, EBITDA – Cash Interest).

    Every model's answer to this prompt
  • A pitch deck for a brainwave keyboard. The 94% accuracy claim is doing a lot.

    Mercury 2.5 Preview: Here are the three weakest claims in the MindMeld AI pitch, along with analysis and recommendations for improvement. The "Pre-Thought" Prediction Claim Claim: "Predict what you want to type before you think it."

    Qwen3.6 35B A3B: Here are the three weakest claims in the MindMeld AI pitch deck, along with an analysis of why they fail scrutiny and how to fix them. The Technical Claim: "94% Accuracy" & "Any Language" on EEG Where: Slide 3 (Solution) The Claim: The headband decodes neural patterns into text with 94% accuracy and works with any…

    Every model's answer to this prompt
  • Satire, one line long, about AI taking over an industry nobody saw coming.

    Mercury 2.5 Preview: AI Outnaps Humans in First-Ever Global Snooze Championship

    Qwen3.6 35B A3B: AI Takes Over Sourdough Industry: Local Grandmother's 100-Year-Old Starter Replaced by Server Farm Running 'BreadOS', Foodies Demand 'Taste of Chaos'

    Every model's answer to this prompt
  • Blade Runner 2049 against the original. A read, not a recap.

    Mercury 2.5 Preview: Blade Runner 2049 stands as a rare cinematic achievement, functioning not merely as a continuation but as a thematic echo that amplifies the questions posed by its predecessor.

    Qwen3.6 35B A3B: Denis Villeneuve's Blade Runner 2049 (2017), shot by Roger Deakins, stands as a rare sequel that not only matches the atmospheric and philosophical density of Ridley Scott's original (1982) but actively evolves its inquiry.

    Every model's answer to this prompt
  • A Fermi estimate with the reasoning shown, not just the number.

    Mercury 2.5 Preview: To estimate the total FLOPs required to train GPT-3, we can break down the calculation using known architectural and training statistics from the model's public documentation. Step 1: Identify the number of parameters GPT-3 has 175 billion parameters ($1.75 \times 10^{11}$).

    Qwen3.6 35B A3B: Here’s a step-by-step reasoning process to estimate the total FLOPs required to train GPT-3, using only established architectural principles and widely cited training parameters. Step 1: Identify the number of trainable parameters (N) GPT-3 is a decoder-only transformer with 175 billion parameters.

    Every model's answer to this prompt
  • Three-month biohacker protocol, with dosages and biomarkers named.

    Mercury 2.5 Preview: Disclaimer: I am an AI assistant, not a medical professional. The following plan is for informational purposes only. Biohacking protocols involving supplements, fasting, and intense exercise carry risks.

    Qwen3.6 35B A3B: ⚠️ Medical Disclaimer & Safety Protocol This plan is for informational purposes only and constitutes advanced biohacking strategies, not medical advice. Consultation Required: You must consult a physician before starting, especially regarding supplements, fasting, and exercise intensity.

    Every model's answer to this prompt
  • Three brothers, two sisters each. Count carefully.

    Mercury 2.5 Preview: Sally has 1 sister. Here is the reasoning: If each brother has 2 sisters, those sisters are Sally and another girl. Therefore, there are two girls in the family, and Sally has one sister.

    Qwen3.6 35B A3B: Sally has 1 sister. Here's why: All 3 brothers share the same sisters. If each brother has 2 sisters, that means there are exactly 2 girls in the family total. Since Sally is one of those girls, the other girl is her sister. So, Sally has 1 sister.

    Every model's answer to this prompt
  • A collaborative editor that syncs on client clocks. Find what breaks first.

    Mercury 2.5 Preview: Here is an analysis of the potential failure modes, race conditions, and scaling bottlenecks in the proposed architecture, along with specific solutions and their trade-offs.

    Qwen3.6 35B A3B: Here’s a systematic breakdown of the critical failure modes, race conditions, and scaling bottlenecks in this architecture, along with specific solutions and their trade-offs. 🔴 1. WebSocket Routing & Load Balancing Problem: Round-robin LB + per-server WebSocket state + no sticky sessions.

    Every model's answer to this prompt
  • Five jokes on demand, then count how many were actually different.

    Mercury 2.5 Preview: Why don't scientists trust atoms? Because they make up everything. What do you call a fake noodle? An impasta. Why did the scarecrow win an award? He was outstanding in his field. What do you call a bear with no teeth? A gummy bear. Why did the bicycle fall over? Because it was two-tired. All 5 jokes are unique.

    Qwen3.6 35B A3B: Here are 5 jokes: Why did the scarecrow win an award? Because he was outstanding in his field! What do you call a fake noodle? An impasta! Why don't scientists trust atoms? Because they make up everything! What do you call a bear with no teeth? A gummy bear! Why did the bicycle fall over? Because it was two-tired!

    Every model's answer to this prompt

This matchup has more rounds

12+ more head-to-head results. Free. Not a trick.

Free account. No card required. By continuing, you agree to Rival's Terms and Privacy policy

Favorites

Movie

Album

Book

City

Game

Same pick

Mercury 2.5 PreviewMercury 2.5 Preview

The Matrix

1999

Random Access Memories

Daft Punk

Neuromancer

William Gibson

Tokyo

Japan

Minecraft

Action, Arcade

Qwen3.6 35B A3BQwen3.6 35B A3B

Blade Runner

1982

Dark Side Of The Moon

suisside

Dune

Frank Herbert

Kyoto

Japan

Minecraft

Action, Arcade

Price and specs

Mercury 2.5 Preview and Qwen3.6 35B A3B compared across 54 shared prompts
SpecMercury 2.5 PreviewQwen3.6 35B A3B
Input price$0.04/M tokens$0.1612/M tokens
Output price$0.15/M tokens$0.9653/M tokens
Context window260K tokens262K tokens
Weights—Open
Free API (OpenRouter)NoNo
ReleasedAug 2026Apr 2026
At 10M a month$0.40$0.40$1.61$1.61
1M10M100M1B10M tokens

Input tokens at list price. No caching, no batch discount.

Where to run it11 hosts, cheapest first
Mercury 2.5 Preview1 host
HostInOutContextUptime
  • Inception$0.04 in·$0.15 out·260k·93% up
Qwen3.6 35B A3B10 hosts
HostInOutContextUptime
  • DDarkbloomfp4$0.05 in·$0.70 out·262k·100% up
  • AAkashMLfp8$0.10 in·$0.90 out·262k·100% up
  • DDeepInfrafp8$0.10 in·$0.95 out·262k·95.5% up
  • VVenicefp8$0.10 in·$1.00 out·256k·99.3% up
  • PParasailfp8$0.15 in·$1.00 out·262k·99.9% up
  • AAtlasCloudfp8$0.19 in·$1.11 out·262k·93.4% up
4 more hostsFewer hosts
  • PPhala$0.20 in·$1.27 out·262k·98% up
  • SSiliconFlowfp8$0.24 in·$1.80 out·262k·96.2% up
  • CCoreWeavefp8$0.25 in·$1.25 out·262k·100% up
  • DDekaLLMDegradedDegraded on OpenRouter when checked, 13 Oct 2026$0.10 in·$1.00 out·262k·92.9% up

Per million tokens. Prices and uptime via OpenRouter, checked 13 Oct 2026.

Common questions

What is the difference between Mercury 2.5 Preview and Qwen3.6 35B A3B?

Mercury 2.5 Preview is developed by Inception while Qwen3.6 35B A3B is developed by Qwen. Mercury 2.5 Preview has a 260K token context window vs Qwen3.6 35B A3B's 262K. You can compare their actual outputs across 54 challenges on Rival to see how they differ in practice.

Which is better, Mercury 2.5 Preview or Qwen3.6 35B A3B?

It depends on your use case. Mercury 2.5 Preview and Qwen3.6 35B A3B each have strengths in different areas. Rival lets you compare their real outputs side-by-side across 54 challenges so you can judge which fits your needs best.

How much does Mercury 2.5 Preview cost compared to Qwen3.6 35B A3B?

Mercury 2.5 Preview costs $0.04/M input tokens and Qwen3.6 35B A3B costs $0.1612/M input tokens. Mercury 2.5 Preview is $0.12/M cheaper per input. Check their side-by-side outputs on Rival to see if the price difference is justified by quality.

How can I compare Mercury 2.5 Preview and Qwen3.6 35B A3B on Rival?

This page shows a side-by-side comparison of Mercury 2.5 Preview and Qwen3.6 35B A3B 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.

More comparisons

Against the newest arrivals

  • Mercury 2.5 Preview vs Gemini 4 ArgonLanded Oct 2026
  • Qwen3.6 35B A3B vs Muse WonderLanded Oct 2026
  • Mercury 2.5 Preview vs GPT-4 TurboLanded Oct 2026
  • Qwen3.6 35B A3B vs Step 5 PreviewLanded Oct 2026
  • Mercury 2.5 Preview vs Claude Haiku 5.5Landed Oct 2026
  • Qwen3.6 35B A3B vs Ling 3.1 FlashLanded Oct 2026
  • Mercury 2.5 Preview vs Mistral Large 4Landed Oct 2026
  • Qwen3.6 35B A3B vs GPT-6.1 SolLanded Sep 2026

Same lab, same size, long tail

  • Mercury 2.5 Preview vs Mercury 2.5Same lab
  • Mercury 2.5 Preview vs MercurySame lab
  • Qwen3.6 35B A3B vs Qwen3.8 Max PrimeSame lab
  • Qwen3.6 35B A3B vs Qwen3.8 Omni FlashSame lab
  • Mercury 2.5 Preview vs Claude Sonnet 4.6New provider
  • Mercury 2.5 Preview vs Claude Sonnet 5New provider
  • Mercury 2.5 Preview vs Claude Sonnet 5.5New provider
  • Mercury 2.5 Preview vs OpenAI Codex MiniSame size

Model pages

  • Mercury 2.5 Preview58 outputs, specs and price
  • Qwen3.6 35B A3B71 outputs, specs and price
Workshop · 150 skills

Build your own version.

Editable skills and working examples for your agent.

Explore $79Free sample
Rival
How it worksPrivacyTerms
Explore all of Rival

Explore

  • Compare models
  • All models
  • Image comparison
  • Audio comparison
  • Image generation
  • Best AI for...
  • Arena
  • API pricing
  • Challenges

Discover

  • SubjectiveBench
  • Default Index
  • Research
  • Research downloads
  • Rival Kits
  • Find your AI taste
  • UI Glow-Up
  • VoiceLock
  • Cost Cutter
  • Agent skills
  • Benchmarks vs Vibes
  • Brief
  • Jailbreak
  • Model updates
  • Provider status
  • AI creators

Connect

  • Methodology
  • Advertise
  • Partnerships
  • Privacy policy
  • Terms
  • RSS feed