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. /
  3. Compare
  4. /
  5. Mercury 2.5vsQwen3 235B A22B Thinking 2507
Updated Sep 2026

Mercury 2.5vsQwen3 235B A22B Thinking 2507

Mercury 2.5 is cheaper than Qwen3 235B A22B Thinking 2507 at $0.04/M vs $0.11/M input tokens.

Mercury 2.5 and Qwen3 235B A22B Thinking 2507 compared across 20 shared prompts
SpecMercury 2.5Qwen3 235B A22B Thinking 2507
Input price$0.04/M tokens$0.11/M tokens
Output price$0.15/M tokens$0.6/M tokens
Context window260K tokens131K tokens
Weights—Open
Free API (OpenRouter)NoNo
ReleasedSep 2026Jul 2025
Side by side

Same prompt, both answers

Opening lines from 6 of the 20 prompts both answered. Formatting stripped, words untouched.

Historical Counterfactual Analysis

If the transistor had been invented in 1920 instead of 1947, trace the likely technological, economic, and geopolitical implications through…

Mercury 2.5660 words
Inventing the transistor in 1920 would have fundamentally reshaped the 20th century, compressing the digital age by roughly two decades. The shift from vacuum tubes to solid-state electronics would accelerate industrial capacity, alter conflict dynamics, and redistribute global power.
Read the full answer
Qwen3 235B A22B Thinking 25072,304 words
Okay, the user is asking about a hypothetical scenario where the transistor was invented in 1920 instead of 1947, and wants me to trace the implications up to 1980. This is a complex counterfactual history question that requires careful analysis of technological, economic, and geopolitical ripple effects.
Read the full answer

Simple 7-Ingredient Recipe

Create a simple recipe for a delicious dish (sweet or savory) using only 7 common pantry ingredients. Include a creative name for the dish…

Mercury 2.5190 words
Pantry Pasta Primavera Prep Time: 5 minutes Cook Time: 15 minutes Servings: 2 Ingredients (7 Common Pantry Items) Pasta (spaghetti, penne, or fusilli) – 8 oz Canned crushed tomatoes – 1 can (14 oz) Olive oil – 3 tablespoons Garlic powder – 1 teaspoon Dried oregano – 1 teaspoon Parmesan cheese (grated) – ½ cup Salt and…
Read the full answer
Qwen3 235B A22B Thinking 2507327 words
🌟 "Pantry Firecracker Cinnamon Crisps" Sweet, crunchy, with a hint of smoky warmth—ready in 15 minutes! Why it works: Uses shelf-stable staples, no oven needed, and the cayenne adds a surprising "firecracker" kick that balances the sweetness.
Read the full answer

Explain Like I'm a Specific Expert

Explain how a large language model (like GPT or Claude) learns and generates text to each of these three audiences: 1. **An experienced…

Mercury 2.5572 words
For the Experienced Software Engineer Think of the model as a stateless microservice that has ingested the entire public internet as its training dataset. During the training phase, it performs a massive distributed optimization job, adjusting internal weights to minimize a loss function based on next-token prediction.
Read the full answer
Qwen3 235B A22B Thinking 2507982 words
For the Experienced Software Engineer (Skeptical of "Next Word Prediction") You've built systems where behavior emerges from state transitions and message passing, so your skepticism about "next token prediction" creating intelligence is understandable—it sounds trivial.
Read the full answer

Startup Pitch Teardown

Here's a pitch deck summary for a startup called "MindMeld AI": **Slide 1 - Vision:** "We're building the future of human-AI collaboration.…

Mercury 2.5544 words
Based on the pitch deck summary, here are the three weakest claims, along with an analysis of their vulnerabilities and concrete suggestions for improvement. The "Pre-Thought" Prediction Claim Claim: "MindMeld AI reads your brainwaves to predict what you want to type before you think it."
Read the full answer
Qwen3 235B A22B Thinking 25071,700 words
Okay, the user wants me to analyze a pitch deck for a startup called MindMeld AI. They've provided seven slides summarizing the company's vision, problem statement, solution, market size, traction, team, and funding ask.
Read the full answer

Beginner's Longevity Plan

Outline a simple, actionable 1-month plan for someone starting their journey towards better health and longevity. Focus on 3 key areas…

Mercury 2.5325 words
Here is a simple, actionable 1-month plan to kickstart your journey toward better health and longevity. This guide focuses on three foundational pillars: Nutrition, Movement, and Sleep. Week 1: Awareness & Small Wins Goal: Build habits without overwhelming yourself.
Read the full answer
Qwen3 235B A22B Thinking 2507682 words
Here’s a simple, no-guilt, beginner-friendly 1-month plan focused on diet, movement, and sleep (the 3 most impactful, science-backed pillars for longevity). Designed for real people with busy lives—no apps, expensive tools, or drastic changes. Goal: Build consistent habits, not perfection. Why These 3 Areas?
Read the full answer

Mini LBO Underwrite

You are underwriting a 5-year LBO for a private equity investment committee. Company (fictional): HarborTech Services (HTS) –…

Mercury 2.5403 words
1. Simplified Year-by-Year LBO Model (FY2026–FY2030) Entry Assumptions: Enterprise Value (EV): $120m × 12.0x = $1,440m Transaction Fees: $1,440m × 2.0% = $28.8m Total Debt at Close: $120m × 5.5x = $660m Term Loan: $120m × 4.0x = $480m Mezzanine: $120m × 1.5x = $180m Equity Required: $1,440m (EV) + $28.8m (Fees) − $660m…
Read the full answer
Qwen3 235B A22B Thinking 2507618 words
HarborTech Services (HTS) LBO Underwriting Analysis All calculations use provided fictional data. Rounded to $0.1m for presentation; full precision used internally.
Read the full answer
Our Verdict
Mercury 2.5
Mercury 2.5
Qwen3 235B A22B Thinking 2507
Qwen3 235B A22B Thinking 2507

Not enough votes to call it. On the specs, nothing separates them.

Mercury 2.5 costs 4.0x less per token.

Too close to call

Reviewing agent-written code?See a Brief PR report

API pricing

Cost per 1M tokens

Mercury 2.5
Input
$0.04
2.8× cheaper
Output
$0.15
4.0× cheaper
Qwen3 235B A22B Thinking 2507
Input
$0.11
Output
$0.60

Mercury 2.5 is cheaper on both: 2.8× input, 4.0× output.

Where to run it

4 hosts, cheapest first

Mercury 2.51 host
HostInOutContextUptime
Inception$0.04 in·$0.15 out·260k·100% up
Qwen3 235B A22B Thinking 25073 hosts
HostInOutContextUptime
Alibaba Cloud$0.23 in·$2.30 out·131k·100% upNNovitafp8$0.30 in·$3.00 out·131k·92.6% upVVenicefp8$0.45 in·$3.50 out·128k·66.5% up

Per million tokens. Prices and uptime via OpenRouter, checked 23 Sep 2026.

Research

What we learned reading every model

FAQ

Common questions

Mercury 2.5 is developed by Inception while Qwen3 235B A22B Thinking 2507 is developed by Qwen. Mercury 2.5 has a 260K token context window vs Qwen3 235B A22B Thinking 2507's 131K. You can compare their actual outputs across 20 challenges on Rival to see how they differ in practice.

It depends on your use case. Mercury 2.5 and Qwen3 235B A22B Thinking 2507 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.

Mercury 2.5 costs $0.04/M input tokens and Qwen3 235B A22B Thinking 2507 costs $0.11/M input tokens. Mercury 2.5 is $0.07/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 Mercury 2.5 and Qwen3 235B A22B Thinking 2507 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.

Keep exploring

More comparisons

Against the newest arrivals

Mercury 2.5 logoDeepSeek V4 Flash Vision Exp logo
Mercury 2.5 vs DeepSeek V4 Flash Vision ExpLanded Sep 2026
Qwen3 235B A22B Thinking 2507 logoSolar Pro 4 logo
Qwen3 235B A22B Thinking 2507 vs Solar Pro 4Landed Sep 2026
Mercury 2.5 logoHy3 logo
Mercury 2.5 vs Hy3Landed Sep 2026
Qwen3 235B A22B Thinking 2507 logoQwen3.7 Flash logo
Qwen3 235B A22B Thinking 2507 vs Qwen3.7 FlashLanded Sep 2026
Mercury 2.5 logoLing 3.0 Flash logo
Mercury 2.5 vs Ling 3.0 FlashLanded Sep 2026
Qwen3 235B A22B Thinking 2507 logoMuse Glimmer 30B logo
Qwen3 235B A22B Thinking 2507 vs Muse Glimmer 30BLanded Sep 2026
Mercury 2.5 logoGLM 5.3 logo
Mercury 2.5 vs GLM 5.3Landed Sep 2026
Qwen3 235B A22B Thinking 2507 logoTernary Bonsai 2 27B logo
Qwen3 235B A22B Thinking 2507 vs Ternary Bonsai 2 27BLanded Sep 2026

Same lab, same size, long tail

Mercury 2.5 logoMercury 2.5 Preview logo
Mercury 2.5 vs Mercury 2.5 PreviewSame lab
Mercury 2.5 logoMercury logo
Mercury 2.5 vs MercurySame lab
Qwen3 235B A22B Thinking 2507 logoQwen3.8 Flash logo
Qwen3 235B A22B Thinking 2507 vs Qwen3.8 FlashSame lab
Qwen3 235B A22B Thinking 2507 logoQwen3.8 Max (0902) logo
Qwen3 235B A22B Thinking 2507 vs Qwen3.8 Max (0902)Same lab
Mercury 2.5 logoHealer Alpha logo
Mercury 2.5 vs Healer AlphaNew provider
Mercury 2.5 logoHorizon Alpha logo
Mercury 2.5 vs Horizon AlphaSame size
Mercury 2.5 logoHorizon Beta logo
Mercury 2.5 vs Horizon BetaSame size
Mercury 2.5 logoHunter Alpha logo
Mercury 2.5 vs Hunter AlphaNew provider

Model pages

Mercury 2.5 logo
Mercury 2.520 outputs, specs and price
Qwen3 235B A22B Thinking 2507 logo
Qwen3 235B A22B Thinking 250759 outputs, specs and price
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