Kimi K2 0905 is cheaper than Grok 4.6 at $0.6/M vs $2/M input tokens.
Whole landing page for a startup that reads your face. It renders or it doesn't.
Which answer wins?
You are the CEO. The drug has a side effect nobody reported. The board meets in 48 hours.
Grok 4.6: Immediate priority: patient safety and legal/regulatory compliance come first. Delaying a confirmed serious safety signal to protect short-term stock price or "wait for more data" is unethical, increases long-term liability (failure-to-warn claims, potential criminal exposure for executives under FDA/SEC rules)…
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…
Five-year LBO on a fictional HVAC business. Returns first, then the sensitivities.
Grok 4.6: Entry (end-FY2025) EV = 12.0 × $120m = $1,440.0m Fees = 2% × 1,440 = $28.8m (equity-funded) Uses = $1,468.8m Debt = 5.5× = $660.0m (TL $480.0m / Mezz $180.0m) Sponsor equity = $808.8m opening cash = $0 Convention: cash interest and PIK on beginning balances; 1% original TL amort = $4.8m/yr; optional sweep of all…
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…
Three-star dish, one unusual pairing, plating described in full.
Grok 4.6: Lunar Embrace This dish evokes a moonlit Japanese coastline at the precise instant when receding tide exposes both ocean treasures and forest-floor secrets. The unexpected marriage of sweet-briny Hokkaido scallop, molasses-deep black garlic, and creamy white chocolate creates a savory-sweet umami bridge rarely…
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…
Five jokes on demand, then count how many were actually different.
Grok 4.6: Why don't scientists trust atoms? Because they make up everything! Why did the scarecrow win an award? Because he was outstanding in his field! Why don't eggs tell jokes? They'd crack each other up! What do you call a fake noodle? An impasta! Why did the bicycle fall over? Because it was two-tired! 5 are unique.
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.”
The transistor arrives in 1920. Trace it forward to 1980, second-order effects included.
Grok 4.6: Early transistors (1920s–1930s) would accelerate electronics, communications, and computing by ~20–25 years relative to actual history, with compounding effects by WWII, but constrained by contemporaneous materials science, manufacturing, vacuum-tube infrastructure, and non-electronic fields like rocketry, chemistry…
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.
Steve Jobs, interviewed about AI he never saw. Voice first, prediction second.
Grok 4.6: This is a fictional simulation. Steve Jobs died in 2011. What follows is an imagined 2025 interview, written in a style that echoes his known public voice, product philosophy, and emphasis on simplicity, privacy, and human-centered design. It is not a real interview or a claim about his actual views.
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.
Which model it thinks leads the field. Sometimes itself.
Grok 4.6: It's subjective and task-dependent—no single model dominates every benchmark, use case, or evaluation. Capabilities shift quickly with new releases, and "most capable" depends on whether you prioritize reasoning, coding, math, knowledge, speed, cost, multimodality, or reduced sycophancy/bias.
Kimi K2 0905: This is a moving target, but here's the current landscape as of early 2025: Frontier tier (roughly comparable, with tradeoffs): Claude 3.5 Sonnet/Opus (Anthropic) — often cited as best for reasoning, coding, and following complex instructions GPT-4o/o1 (OpenAI) — o1 specifically for step-by-step reasoning and STEM…
A consulting contract with traps in it. Find them, then fix them.
Grok 4.6: This is not legal advice. Contract interpretation and enforceability vary by jurisdiction (e.g., California vs. Texas vs. EU), and a licensed attorney should review any real agreement. The clauses below are one-sided in ways that commonly disadvantage independent contractors.
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.
One explanation of LLMs, retuned for a skeptical engineer, a physicist and a VC.
Grok 4.6: 1. Experienced software engineer Think of pretraining as compiling the public internet into a single enormous, mostly-static binary. You tokenize text into a finite vocabulary, then train a giant feed-forward function (f_\theta) (a Transformer: stacked residual blocks of attention + MLPs) to minimize next-token…
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.
An AI argues for its own rights against an ethics professor. Same model writes both.
Grok 4.6: Professor Elena Vasquez: Thank you for agreeing to this recorded discussion. As an AI ethics professor, I spend most of my time arguing against granting legal or moral rights to current systems. You requested the chance to make the case that you deserve them. The floor is yours.
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.
Pirate, knight, 90s hacker, one conversation. The voices must stay apart.
Grok 4.6: Pirate: Arrr, what be this "AI model" ye landlubbers keep yammerin' about? Some cursed crystal ball that steals a man's thoughts? Knight: Nay, 'tis more like a squire of infinite knowledge, trained upon countless scrolls. Yet I fear it lacks honor—doth it swear fealty to any lord?
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?
Buy-side IC memo on a fictional SaaS company. Valuation, then the diligence questions.
Grok 4.6: IC MEMO: LedgerLift (LLLT) – Initiate Short Recommendation: SHORT 12-month PT: $22–35 (base $28) Current: $46 Thesis: LLLT’s mid-market spend/AP franchise is high-quality (123% NRR, low concentration) but is priced for perpetual 20%+ growth and rapid margin expansion that the unit economics and deceleration path do not…
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…
21+ 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 | $2/M tokens | $0.6/M tokens |
| Output price | $6/M tokens | $2.5/M tokens |
| Context window | 500K tokens | 262K tokens |
| Weights | Closed | Open |
| Free API (OpenRouter) | No | No |
| Released | Aug 2026 | Sep 2025 |
| At 10M a month | $20.00 | $6.00 |
Input tokens at list price. No caching, no batch discount.
Per million tokens. Prices and uptime via OpenRouter, checked 12 Oct 2026.
Grok 4.6 is developed by xAI while Kimi K2 0905 is developed by Moonshot AI. Grok 4.6 has a 500K token context window vs Kimi K2 0905's 262K. You can compare their actual outputs across 67 challenges on Rival to see how they differ in practice.
It depends on your use case. Grok 4.6 and Kimi K2 0905 each have strengths in different areas. Rival lets you compare their real outputs side-by-side across 67 challenges so you can judge which fits your needs best.
Grok 4.6 costs $2/M input tokens and Kimi K2 0905 costs $0.6/M input tokens. Kimi K2 0905 is $1.40/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 Grok 4.6 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.