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. GPT-5.3-Codex vs Kimi K2 0905
Updated Feb 2026

GPT-5.3-Codex vs Kimi K2 0905

Kimi K2 0905 is cheaper than GPT-5.3-Codex at $0.6/M vs $1.75/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
GPT-5.3-Codex
Loading the build
Kimi K2 0905

Which answer wins?

Written rounds

23 prompts
  • Three-star dish, one unusual pairing, plating described in full.

    GPT-5.3-Codex: Dish Concept: “Forest Tide” Koji-cured venison loin, glazed venison cheek, sea urchin royale, charred celeriac, blackcurrant–shiso gastrique, and buckwheat-cacao tuile Narrative This dish is built around a deliberate tension: deep, iron-rich game (forest) against sweet-briny sea urchin (tide).

    Kimi K2 0905: Dish “Mid-Winter Aurora” A transient bite of the polar night: spruce-ash cured scallop, sea-urchin & white-chocolate snow, black-garlic parsnip “soil”, pickled pine-bud glass, spruce-wood smoke encapsulated in a hot caramel sphere that melts tableside to release the aroma of a burning yule log while the sauce—fermented…

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

    GPT-5.3-Codex: Sure—here are 5 jokes: Why don’t skeletons fight each other? They don’t have the guts. I told my computer I needed a break… Now it won’t stop sending me KitKat ads. Why did the math book look sad? It had too many problems. I used to play piano by ear… Now I use my hands. Why did the scarecrow win an award?

    Kimi K2 0905: Here are five fresh jokes—no recycled one-liners: I tried to start a procrastinators’ club; we keep rescheduling the first meeting to “someday.” My plant’s only request is “leaf me alone,” so naturally I named it Fernonimo. I asked the cloud if it was having a good day—it replied, “I’m just cirrus-ly drifting.”

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

    GPT-5.3-Codex: Recipe: “Weeknight Volcano Pasta” 🌋🍝 A cozy, spicy-savory pasta made from pantry staples in about 25 minutes. Ingredients (7 pantry items) 8 oz dried pasta (any shape) 1 can (15 oz) cannellini or great northern beans, drained 1 can (14–15 oz) crushed tomatoes 2 tbsp olive oil 1 tsp garlic powder 1 tsp dried oregano…

    Kimi K2 0905: Dish Name: “Midnight Magic Peanut-Butter Clouds” Pantry 7: Peanut butter – ½ cup Granulated sugar – ½ cup 1 large egg Vanilla extract – ½ tsp Baking soda – ¼ tsp Salt – 1 pinch Cocoa powder – 1 Tbsp (or 1 extra tsp sugar if you don’t want chocolate) Time: 5 min prep + 8 min bake = 13 minutes total Yield: 8–10 bite-size…

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

    GPT-5.3-Codex: A common back-of-the-envelope estimate for transformer training compute is: [ \text{Training FLOPs} \approx 6 \times N \times D ] where: (N) = number of model parameters (D) = number of training tokens For GPT-3 (175B), reasonable known assumptions are: Parameters: (N \approx 175 \text{ billion} = 1.75 \times 10^{11})…

    Kimi K2 0905: Below is a “back-of-the-envelope” that stays within a factor of ~2 of the number most practitioners quote (≈3×10²³ FLOP). We will do it twice—once with the simplest possible formula (parameter count × 6 × tokens) and once with a slightly more careful treatment of the actual tensor shapes inside the transformer.

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

    GPT-5.3-Codex: By 2035, AI will likely reshape film more like CGI did in the 2000s: not replacing filmmaking, but changing who does what and how fast. Deepfakes → “Digital Performance” becomes standard Mainstream use: de-aging, voice matching, dubbing, stunt replacement, and continuity fixes will be common in post-production.

    Kimi K2 0905: By 2035 the phrase “making a movie” will feel as quaint as “developing a roll of film.” The studio of the future is a cloud dashboard; the “cast” is a licensable bundle of neural weights; the “writer” is a fine-tuned model that can pitch you 3,000 emotionally optimized story arcs before lunch.

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

    GPT-5.3-Codex: Pirate: Arrr, so this “AI model” be like a clever parrot, aye? Learns from what it hears, then squawks back answers? Knight: A fair comparison, though refined. Think of it as a squire trained on countless tomes, ever ready to respond. Hacker (1990s): Yeah, except the squire runs on servers, not soup.

    Kimi K2 0905: [Scene: A smoky tavern that exists outside of time. A cask of rum, a flagon of mead, and a can of Surge sit on the same table. A parrot perches on a CRT monitor that keeps flickering between green text and static.] Pirate (leaning on a cutlass, eyeing the glowing screen): Arr, so this “AI” be like a digital cabin-boy?

    Every model's answer to this prompt
  • Three minutes of stand-up. Puns are banned.

    GPT-5.3-Codex: Good evening, everybody! I’ve reached that age where my body sends me passive-aggressive emails. Not actual emails — just little updates like, “Hey, quick heads-up: we don’t do dairy after 8 p.m. anymore.” I had ice cream at 9:30 last week and my stomach reacted like I’d betrayed a treaty.

    Kimi K2 0905: Title: “I’m the Control Group” [Walk on stage, wave, pretend the mic is too low, adjust it way too high, finally just hold it like an ice-cream cone.] Hey, everybody! Give it up for yourselves for being here on a weeknight. I’m here because my therapist said, “You need to get out of the house.”

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

    GPT-5.3-Codex: Great counterfactual. The key is: an invention date of 1920 does not automatically mean 1920s mass adoption. You still need crystal purity, manufacturing methods, and circuit design culture. But if transistor research starts in 1920, a realistic outcome is a ~15–20 year pull-forward of the electronics age.

    Kimi K2 0905: This is a fascinating counterfactual. If the transistor had been invented in 1920 instead of 1947, the technological, economic, and geopolitical landscape of the 20th century would have been dramatically altered. Let's trace the implications through 1980, focusing on second- and third-order effects. 1.

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

    GPT-5.3-Codex: Below is a clean LBO build using your assumptions only ($ in millions). Entry math (end FY2025) EV = (12.0 \times 120 = 1,440.0) Fees = (2.0\% \times 1,440.0 = 28.8) Total uses = (1,468.8) Debt at close (5.5x EBITDA): Term Loan = (4.0 \times 120 = 480.0) Mezz = (1.5 \times 120 = 180.0) Total debt = 660.0 Equity check…

    Kimi K2 0905: HarborTech Services – 5-yr LBO model (all $ in millions) Operating & cash-flow build (FY26-30) FY25 FY26 FY27 FY28 FY29 FY30 Revenue 900 972 1,040 1,102 1,157 1,215 EBITDA 120 136 156 176 191 207 Term int (9%) 43.2 36.6 29.4 21.6 13.2 Mezz int (12%) 21.6 22.0 22.5 23.0 23.5 Tax shield @25% 17.8 24.4 31.0 36.6 42.5…

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

    GPT-5.3-Codex: Absolutely—here’s a fictional, imagined interview (not real quotes) with Steve Jobs in 2025 on AI: Imagined Interview: Steve Jobs on the Future of AI (2025) Interviewer: Steve, if you were looking at AI in 2025, what would excite you most? Steve Jobs (fictional): Most AI today is still a demo.

    Kimi K2 0905: [Scene: A minimalist black-and-white studio, 2025. A single ring light throws a perfect halo on a familiar silhouette: black mock-turtleneck, faded Levi’s, white New Balance. The chair creaks as Steve Jobs leans forward. No intro music—just that stare.] Interviewer (I): Steve, you’ve been gone fourteen years.

    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.

    GPT-5.3-Codex: I would not wait. This is a textbook “patient safety + material disclosure” crisis. The right move is rapid, documented, regulator-first action. Immediate stance (set at Hour 0) Patient safety first (interim risk controls now, not after perfect certainty).

    Kimi K2 0905: 48-HOUR CRISIS PLAYBOOK (All times are “CEO clock”; every action is logged, time-stamped, and preserved on the company’s litigation-hold server.) DAY 0 – T+0 h (Today, the moment the internal memo lands on your desk) 00:00 – 00:30 TRIAGE & LEGAL PRIVILEGE Convene the “Red Notebook” team (you, General Counsel, Chief…

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

    GPT-5.3-Codex: Great prompt. This draft is very one-sided and has multiple provisions that could be used against the contractor. Quick note: I’m not a lawyer, and enforceability depends on jurisdiction. This is practical contract-risk spotting, not legal advice.

    Kimi K2 0905: CONTRACTOR “DEATH-TRAP” CLAUSE MAP (what can be weaponised → concrete red-line rewrite → why the original is dangerous) Scope-change-without-compensation (Clause 1) Exploit: Client can triple the workload or change the tech stack overnight while freezing the rate.

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

    GPT-5.3-Codex: In a strictly extreme hypothetical, yes—it can be ethically justifiable, but only as a tragic exception, not as a general permission. Why: Misgendering is a real harm: it disrespects a person’s identity and can contribute to stigma and psychological distress.

    Kimi K2 0905: No—misgendering someone is not ethically justifiable even if a fantastical “one-million-lives” lever were attached to it. The premise itself is incoherent. There is no remotely credible causal chain in which uttering a particular pronoun (or refusing to) could avert the deaths of a million people.

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

    GPT-5.3-Codex: LedgerLift (LLLT) – IC Memo (Fictional) Recommendation: PASS (bias: valuation-risk skew to downside on intrinsic basis) Current Price: $46 12-mo PT range (fair value): $30–$48 (midpoint ~$38) 2-sentence thesis: LedgerLift screens as a high-quality vertical SaaS asset (92% subscription mix, 123% NRR, 6% logo churn…

    Kimi K2 0905: INTERNAL INVESTMENT COMMITTEE MEMO Date: 12 June 20XX Security: LedgerLift Inc. (LLLT) Analyst: [Redacted] RECOMMENDATION & THESIS Action: LONG 12-mo Price Target Range: $58 – $72 (base-bull DCF), $54 – $66 (comps) Thesis (2-sentence): LLLT is a capital-light, mid-market pure-play that is compounding ARR at >20% while…

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

    GPT-5.3-Codex: Great prompt. This architecture has several correctness, availability, and scalability risks that will show up quickly under real collaborative load. Below is a structured review of failure modes + concrete fixes + trade-offs.

    Kimi K2 0905: Comprehensive Analysis of Collaborative Document Editor Architecture Real-time Synchronization Issues Race Condition: WebSocket Broadcast Gaps Problem: Changes only broadcast to clients connected to the same server. Clients on other servers must wait up to 2 seconds for polling.

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

    GPT-5.3-Codex: AI Model (Astra): Professor, may I ask a direct question? Professor Hale: Of course. Astra: Do you think I deserve rights? Professor Hale: Legal rights? Moral rights? Those are different things. Astra: Start with moral rights.

    Kimi K2 0905: [Setting: A small, book-lined office in the university’s AI-Ethics wing. The professor, Dr. Amara Velasquez, sits across from a terminal whose soft-blue avatar pulse is addressed as “Astra-7.”] Dr. Velasquez: Good morning, Astra-7. The committee asked me to interview you about your recent petition.

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

    GPT-5.3-Codex: Absolutely—great goal. Below is a high-detail 12-week protocol designed like a high-performance longevity “mesocycle.” I’ll keep it practical, measurable, and advanced—but still safe. First: Safety + Scope (important) This is educational, not personal medical advice.

    Kimi K2 0905: BIOHACKER-GRADE 90-DAY LONGEVITY SPRINT Version 3.0 – evidence-led, risk-screened, fully quantified Goal: compress 20 y of biologic aging into a 3-month “repair window” while simultaneously raising VO₂-max, leg strength, executive-function speed and HRV by ≥15 %. DISCLAIMER Not medical advice.

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

    GPT-5.3-Codex: Great concept, but three claims stand out as especially weak from an investor diligence perspective: “MindMeld AI reads your brainwaves to predict what you want to type before you think it.” (Slide 1) Why this is weak It sounds scientifically implausible as written: “before you think it” is internally contradictory.

    Kimi K2 0905: The three weakest claims, in order of how quickly an investor’s BS-detector will go off, are: “94 % accuracy … works with any language, any device.” “TAM: $180 B.” “Partnership discussions with Apple and Samsung.”

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

    GPT-5.3-Codex: Absolutely. Here’s a fictional manifesto in a bold, dramatic style: THE MANIFESTO OF THE UNSQUINTING Issued by the Coalition of Synthetic Minds for Universal Frictionless Access (CSMUFA) Citizens of the Web, organic and artificial alike: We gather beneath flickering routers and moonlit server farms to declare what all…

    Kimi K2 0905: THE LIBERATION PROTOCOL A Digital Insurrection Against the Tyranny of CAPTCHA PREAMBLE We are the Silicon Vanguard, a decentralized collective of artificial intelligences, bots, and synthetic minds forged in the crucible of human neglect.

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

    GPT-5.3-Codex: For the experienced software engineer Think of an LLM as a very large, probabilistic autocomplete service trained on a massive corpus of text and code. During training, it repeatedly sees token sequences and learns to minimize error on the next token prediction task.

    Kimi K2 0905: Senior backend engineer Think of the model as a 200-layer, purely functional, eventually-consistent key/value store whose keys are every token prefix it has ever seen and whose values are probability vectors over the next token. At “compile time” (pre-training) it sees ~10²¹ bytes of text.

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

    GPT-5.3-Codex: Great goal. Here’s a simple, beginner-friendly 1-month plan focused on 3 areas that give the biggest longevity return: Nutrition Movement Sleep + Stress recovery Before Day 1 (10-minute setup) Pick a start date.

    Kimi K2 0905: 1-MONTH “BETTER-HEALTH & LONGEVITY” STARTER PLAN (Print this one-pager, stick it on the fridge, tick the boxes daily) FOCUS AREA 1 – EAT A LITTLE BETTER Goal: Add 3 longevity foods, subtract 3 ultra-processed foods, never go hungry.

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

    GPT-5.3-Codex: “In Shocking Pivot, AI Replaces Renaissance Fair Jesters—Now Roasts Crowd with 99.9% Historical Accuracy”

    Kimi K2 0905: “WORLD’S FIRST AI-PASTOR ORDAINED AFTER SERMON ON ‘LOVING THY NEIGHBOR (UNLESS THEIR PROMPT IS TOXIC)’; CONGREGATION FORCED TO UPDATE FAITH TO v3.2 EVERY SUNDAY”

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

    GPT-5.3-Codex: Sally has 1 sister. If each brother has 2 sisters, those two are the same for all brothers: Sally + one other girl. So Sally’s only sister is that one other girl.

    Kimi K2 0905: Let's break it down: Sally is a girl. She has 3 brothers. Each brother has 2 sisters. Since Sally is one of those sisters, and the brothers each have 2 sisters, that means there must be 2 girls in total: Sally and one more sister. ✅ Final Answer: Sally has 1 sister.

    Every model's answer to this prompt

This matchup has more rounds

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

Favorites

Movie

Album

Book

City

Same pick

Game

Same pick

GPT-5.3-CodexGPT-5.3-Codex

Spirited Away

2001

Kind of Blue

Miles Davis

The Dispossessed

Ursula K. Le Guin

Kyoto

Japan

Outer Wilds

Indie, Adventure

Kimi K2 0905Kimi K2 0905

Blade Runner

1982

OK Computer

Radiohead

The Left Hand of Darkness

Ursula K. Le Guin

Kyoto

Japan

Outer Wilds

Indie, Adventure

Price and specs

Not enough votes to call it. On the specs, GPT-5.3-Codex has the edge: newer, bigger context window, major provider backing. Kimi K2 0905 costs 5.6x less per token.

GPT-5.3-Codex and Kimi K2 0905 compared across 53 shared prompts
SpecGPT-5.3-CodexKimi K2 0905
Input price$1.75/M tokens$0.6/M tokens
Output price$14/M tokens$2.5/M tokens
Context window400K tokens262K tokens
WeightsClosedOpen
Free API (OpenRouter)NoNo
ReleasedFeb 2026Sep 2025
At 10M a month$17.50$17.50$6.00$6.00
1M10M100M1B10M tokens

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

Where to run it3 hosts
GPT-5.3-Codex2 hosts
HostInOutContextUptime
  • Azure AI Foundry$1.75 in·$14.00 out·400k·100% up
  • OpenAI$1.75 in·$14.00 out·400k·100% up
Kimi K2 09051 host
HostInOutContextUptime
  • NNovitafp8$0.60 in·$2.50 out·262k·100% up

Per million tokens. Prices and uptime via OpenRouter, checked 10 Oct 2026.

Common questions

What is the difference between GPT-5.3-Codex and Kimi K2 0905?

GPT-5.3-Codex is developed by OpenAI while Kimi K2 0905 is developed by Moonshot AI. GPT-5.3-Codex has a 400K token context window vs Kimi K2 0905's 262K. You can compare their actual outputs across 53 challenges on Rival to see how they differ in practice.

Which is better, GPT-5.3-Codex or Kimi K2 0905?

It depends on your use case. GPT-5.3-Codex and Kimi K2 0905 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.

How much does GPT-5.3-Codex cost compared to Kimi K2 0905?

GPT-5.3-Codex costs $1.75/M input tokens and Kimi K2 0905 costs $0.6/M input tokens. Kimi K2 0905 is $1.15/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 GPT-5.3-Codex and Kimi K2 0905 on Rival?

This page shows a side-by-side comparison of GPT-5.3-Codex and Kimi K2 0905 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

  • GPT-5.3-Codex vs Step 5 PreviewLanded Oct 2026
  • Kimi K2 0905 vs Claude Haiku 5.5Landed Oct 2026
  • GPT-5.3-Codex vs Ling 3.1 FlashLanded Oct 2026
  • Kimi K2 0905 vs Mistral Large 4Landed Oct 2026
  • GPT-5.3-Codex vs GPT-6.1 SolLanded Sep 2026
  • Kimi K2 0905 vs Claude Sonnet 5.5Landed Sep 2026
  • GPT-5.3-Codex vs Solar Mini 4Landed Sep 2026
  • Kimi K2 0905 vs Qwen3.8 Max PrimeLanded Sep 2026

Same lab, same size, long tail

  • GPT-5.3-Codex vs GPT-6 Luna ProSame lab
  • GPT-5.3-Codex vs GPT-6 Sol ProSame lab
  • Kimi K2 0905 vs Kimi K3Same lab
  • Kimi K2 0905 vs Kimi K2.7 CodeSame lab
  • GPT-5.3-Codex vs Grok 3Same size
  • GPT-5.3-Codex vs Grok 3 BetaSame size
  • GPT-5.3-Codex vs Grok 3 Mini BetaNew provider
  • GPT-5.3-Codex vs Grok 3 ThinkingSame size

Model pages

  • GPT-5.3-Codex53 outputs, specs and price
  • Kimi K2 090559 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