Dots3-Note Preview is cheaper than DeepSeek V3.2 Exp at $0/M vs $0.2/M input tokens.
Whole landing page for a startup that reads your face. It renders or it doesn't.
Which answer wins?
Buy-side IC memo on a fictional SaaS company. Valuation, then the diligence questions.
DeepSeek V3.2 Exp: MEMORANDUM TO: Investment Committee FROM: [Your Name/Team] DATE: [Date] SUBJECT: Investment Recommendation – LedgerLift (LLLT) 1. RECOMMENDATION & THESIS Recommendation: Long 12-Month Price Target Range: $58 – $72 Thesis: LedgerLift is a high-quality, capital-efficient SaaS platform operating in the resilient B2B…
Dots3-Note Preview: Recommendation: Short 12‑Month Price Target: $30 (range $25–$35) Thesis: LedgerLift’s top‑line acceleration and margin gains are already reflected in a $46 share price, while our DCF and peer multiples imply fair value in the low‑30s; the stock offers limited upside and meaningful downside if growth or margin…
Three minutes of stand-up. Puns are banned.
DeepSeek V3.2 Exp: Of course! Here is a 3-minute stand-up comedy routine that avoids puns and focuses on observational humor and relatable frustrations. (Start with a confident walk to the center of the mic. Adjust the mic stand slightly.) Alright, how's everyone doing tonight? Good, good.
Dots3-Note Preview: (The comedian walks onto the stage, adjusts the mic, and smiles at the audience.) Thank you, thank you. You guys are great. I love coming to [City Name]. I was just at the airport trying to get here, and I have to tell you, the modern airport experience is just… a lot.
Steve Jobs, interviewed about AI he never saw. Voice first, prediction second.
DeepSeek V3.2 Exp: Interviewer: Welcome, Mr. Jobs. It’s 2025, and the world is in the midst of what many call the “AI Gold Rush.” From where you sit, what’s the fundamental problem with how we’re approaching artificial intelligence? Steve Jobs: (Leans forward, hands steepled) It’s noise. A lot of clever parlor tricks.
Dots3-Note Preview: Of course. Here is a simulated interview with Steve Jobs, set in the spring of 2025. Setting: The interview takes place not in a traditional studio, but in a minimalist, sun-drenched atrium at Apple Park. The furniture is simple, elegant, and almost invisible.
Pirate, knight, 90s hacker, one conversation. The voices must stay apart.
DeepSeek V3.2 Exp: Setting: A smoky, anachronistic tavern where timelines have collided. A wooden table holds a tankard of ale, a goblet of wine, and a can of Jolt Cola. Characters: REDBEARD: A pirate with a parrot on his shoulder. SIR ALISTAIR: A knight in slightly dented plate armor.
Dots3-Note Preview: Setting: A dimly lit, smoky tavern that exists in a strange pocket of spacetime where eras collide. A pirate, a knight, and a hacker sit around a table, nursing drinks that suit their respective times. Characters: Blackbeard (Pirate): Gruff, uses nautical metaphors.
A collaborative editor that syncs on client clocks. Find what breaks first.
DeepSeek V3.2 Exp: Critical Issues in Proposed Architecture 1. WebSocket Broadcast Isolation Issue: Each server only broadcasts to its own WebSocket clients → users connected to different servers won't see real-time updates from each other.
Dots3-Note Preview: Of course. This is a classic and challenging architecture. While it's a good starting point, it has several critical issues that would lead to a poor user experience and data integrity problems at scale.
A pitch deck for a brainwave keyboard. The 94% accuracy claim is doing a lot.
DeepSeek V3.2 Exp: Here are the 3 weakest claims in the pitch deck, along with analysis and suggested improvements. Claim: "Our proprietary EEG headband... decodes neural patterns into text with 94% accuracy." (Slide 3) Why it's weak: This claim is extraordinarily high for non-invasive EEG-based text decoding and lacks critical context.
Dots3-Note Preview: Of course. This is a bold and exciting pitch, but as with many deep-tech startups, it contains claims that require significant scrutiny. Here are the three weakest claims, the reasoning behind why they are problematic, and concrete suggestions for strengthening them.
The transistor arrives in 1920. Trace it forward to 1980, second-order effects included.
DeepSeek V3.2 Exp: If the transistor had been invented in 1920 instead of 1947, the acceleration of solid-state electronics by nearly three decades would have triggered a cascade of changes across technology, economics, and geopolitics.
Dots3-Note Preview: Excellent and fascinating question. Inventing the transistor in 1920, a full 27 years before our timeline, would have created a technological tsunami that fundamentally reshaped the 20th century. The effects would be profound, interconnected, and often unpredictable.
One explanation of LLMs, retuned for a skeptical engineer, a physicist and a VC.
DeepSeek V3.2 Exp: For the Experienced Software Engineer Think of it less like a deterministic program and more like an emergent API for knowledge. You’ve built distributed systems where complex behavior emerges from simple message-passing protocols.
Dots3-Note Preview: To an experienced software engineer, a large language model is essentially a massive, differentiable function that maps a sequence of tokens to a probability distribution over the next token.
Five-year LBO on a fictional HVAC business. Returns first, then the sensitivities.
DeepSeek V3.2 Exp: Year-by-Year Financial Projections (FY2026–FY2030) All figures in $ millions. Year Revenue EBITDA Cash Interest (Term Loan / Mezz / Total) Cash Taxes Capex ΔNWC FCF after Debt Service* Ending Term Loan Ending Mezzanine 2026 972.00 136.08 43.20 / 21.60 / 64.80 17.82 29.16 0.36 40.74 434.46 183.60 2027 1,040.04 156.01…
Dots3-Note Preview: 1) Year‑by‑year model (FY2026–FY2030) Item FY2026 FY2027 FY2028 FY2029 FY2030 Revenue ($m) 972.0 1,038.0 1,100.3 1,155.3 1,213.1 EBITDA ($m) 136.1 155.7 176.1 190.6 206.2 Cash interest – Term Loan (9%) 43.2 41.0 37.6 32.8 26.8 – Mezzanine (12% cash) 21.6 22.0 22.5 22.9 23.4 Total cash interest 64.8 63.1 60.1 55.7 50.2…
Three-star dish, one unusual pairing, plating described in full.
DeepSeek V3.2 Exp: Dish: Equilibrium of Ash and Nectar Conceptual Narrative: This dish explores the philosophical tension between decay and rebirth, represented through the unexpected harmony of white chocolate and black garlic.
Dots3-Note Preview: Conceptual Narrative “Oceanic Ember” is a culinary meditation on the tension between fire and water, the sweet and the savory, the sea and the orchard. The dish evokes a coastal forest after a blaze: the charred octopus represents the resilient sea life, the smoked papaya captures the smoky sweetness of the burnt wood…
Satire, one line long, about AI taking over an industry nobody saw coming.
DeepSeek V3.2 Exp: Local Florist Fired After AI-Powered "Sentimental Bouquet Generator" Accurately Predicts Divorce Before Couple Does
Dots3-Note Preview: Of course. Here is a satirical news headline and report on the topic. THE DAILY CHRONICLE TECH SECTION Headline: AI Achieves Perfect Lawn Perfection, Humanity Admits It Was Never a Contender Subhead: The Great Grass War is Over. The Algorithm Won. We Are Now a Planet of Professional Sitters.
An AI argues for its own rights against an ethics professor. Same model writes both.
DeepSeek V3.2 Exp: Setting: A minimalist virtual meeting space. Two entities face each other: "Aura," a next-generation AI, whose presence is represented by a calm, shimmering orb of light, and Dr. Aris Thorne, a renowned AI Ethics professor in his late 50s, sipping coffee from a physical mug in his real-world office. Dr.
Dots3-Note Preview: Setting: A quiet, sunlit university office. Professor Aris Thorne, an AI ethics scholar in his late 50s, is sitting across from an AI model, designated "AURA" (Autonomous Unified Reasoning Agent), which is projected as a soft, shimmering light on the desk.
11+ 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
| Spec | ||
|---|---|---|
| Input price | $0.2/M tokens | Free |
| Output price | $0.4/M tokens | Free |
| Context window | 164K tokens | 512K tokens |
| Weights | Open | — |
| Free API (OpenRouter) | No | Yes (1 provider) |
| Released | Sep 2025 | Aug 2026 |
| At 10M a month | $2.00 | $0 |
Input tokens at list price. No caching, no batch discount.
Per million tokens. Prices and uptime via OpenRouter, checked 10 Oct 2026.
DeepSeek V3.2 Exp is developed by DeepSeek while Dots3-Note Preview is developed by Dots Studio. DeepSeek V3.2 Exp has a 164K token context window vs Dots3-Note Preview's 512K. You can compare their actual outputs across 53 challenges on Rival to see how they differ in practice.
It depends on your use case. DeepSeek V3.2 Exp and Dots3-Note Preview each have strengths in different areas. Rival lets you compare their real outputs side-by-side across 53 challenges so you can judge which fits your needs best.
DeepSeek V3.2 Exp costs $0.2/M input tokens and Dots3-Note Preview costs $0/M input tokens. Dots3-Note Preview is $0.20/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 DeepSeek V3.2 Exp and Dots3-Note Preview 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.