Skip to content
llmwise
Sign inStart free

Kimi K3 · Math

Kimi K3 for math: our test runs

Kimi K3 ran our 5 math prompts in llmwise, through the same pipeline your messages take. Here's every reply as it came, whether it passed, what it cost and what it counts as on Pro, next to the other models.

Start freeSee pricing

Based on 5 of our test runs on , through OpenRouter with the app's own prompt and settings. Updated .

Short answer

In our test runs on September 27, 2026, Kimi K3 passed 5 of 5 of our math prompts (both hard ones): tied 1st of 20 models. Each reply counted as 1 message on Pro, where it has 125 a month, and cost $0.0037 to run on average.

Prompt by prompt

Each of the 5 math prompts, and how Kimi K3's reply did.

Kimi K3 on each math prompt
PromptResultCostTime
A discount, then sales taxEverydayPassed: Final answer 66: right.$0.00375.0 s
Pens at 3 for $4EverydayPassed: Final answer Buy three 3-for- 4 packs and one single pen; 13.50: right.$0.00465.4 s
Compound interest over three yearsEverydayPassed: Final answer 315.25: right.$0.00445.9 s
Four-digit numbers whose digits sum to 9HardPassed: Final answer 165: right.$0.00326.0 s
The highest of three dice is a 5HardPassed: Final answer 61/216: right.$0.00272.4 s

Every math reply from Kimi K3

  1. A discount, then sales tax

    An everyday prompt: “A discount, then sales tax”, in full.

    Passed: Final answer 66: right.

    Discounted price: $80 × 0.75 = $60  
Add 10% tax: $60 × 1.10 = $66  

Final answer: $66

    489 tokens in, 125 out (75 of them reasoning) · 5.0 s · $0.0037 · 1 message on Pro · answered by moonshotai/kimi-k3 via Fireworks ·

  2. Pens at 3 for $4

    An everyday prompt: “Pens at 3 for $4”, in full.

    Passed: Final answer Buy three 3-for- 4 packs and one single pen; 13.50: right.

    Three 3-for-$4 packs give 9 pens for $12; add 1 single for $1.50 to reach exactly 10. Total = $12 + $1.50 = $13.50.

Final answer: Buy three 3-for-$4 packs and one single pen; $13.50

    489 tokens in, 322 out (247 of them reasoning) · 5.4 s · $0.0046 · 1 message on Pro · answered by moonshotai/kimi-k3 via Phala ·

  3. Compound interest over three years

    An everyday prompt: “Compound interest over three years”, in full.

    Passed: Final answer 315.25: right.

    Amount after 3 years: \(2000 \times 1.05^3 = 2315.25\)  
Interest earned: \(2315.25 - 2000 = 315.25\)

Final answer: $315.25

    491 tokens in, 171 out (107 of them reasoning) · 5.9 s · $0.0044 · 1 message on Pro · answered by moonshotai/kimi-k3 via Fireworks ·

  4. Four-digit numbers whose digits sum to 9

    A hard prompt: “Four-digit numbers whose digits sum to 9”, in full.

    Passed: Final answer 165: right.

    Let the digits be \(a,b,c,d\) with \(a\ge1\) and \(a+b+c+d=9\).

Set \(a'=a-1\), so \(a'\ge0\) and:

\[
a'+b+c+d=8
\]

Since the total is 8, no digit can exceed 9, so upper limits don’t matter. By stars and bars:

\[
\binom{8+4-1}{4-1}=\binom{11}{3}=165
\]

Final answer: 165

    465 tokens in, 332 out (210 of them reasoning) · 6.0 s · $0.0032 · 1 message on Pro · answered by moonshotai/kimi-k3 via Wafer ·

  5. The highest of three dice is a 5

    A hard prompt: “The highest of three dice is a 5”, in full.

    Passed: Final answer 61/216: right.

    Count outcomes where all dice are at most 5: \(5^3=125\).  
Subtract outcomes where all dice are at most 4: \(4^3=64\).  
So exactly highest 5: \(125-64=61\) out of \(6^3=216\).

Final answer: 61/216

    480 tokens in, 176 out (96 of them reasoning) · 2.4 s · $0.0027 · 1 message on Pro · answered by moonshotai/kimi-k3 via Phala ·

How it compares on math

Kimi K3 beside the job's picks among every model. 19 models share the hard-problems pick, Kimi K3 among them.

Kimi K3 beside other models on math
ModelPassedOn ProCost per reply
GLM 5.3Our pick: hard problems (shared), best value5 of 5250 a month on Pro$0.0004
Claude Haiku 5.5Our pick: hard problems (shared), everyday5 of 560 a day on Pro$0.0001
Kimi K3Our pick: hard problems (shared)5 of 5125 a month on Pro$0.0037
On Pro: Pro's count on each model. Cost: what OpenRouter charged us per reply, on average.

Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Plus $3.75, Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.

Kimi K3 next to each model on math

Kimi K3 against each model on the same 5 math prompts. One run each, through OpenRouter: a wait depends on the provider and the load that day, so a lead under 10% counts as close.

  • Against Claude Fable 5.1: Kimi K3 answered 1.2× later (5.4 s to 4.6 s), cost 3.1× less, and passed 5 of 5 to its 5.

  • Against Claude Opus 5.5: Kimi K3 answered 1.4× later (5.4 s to 3.9 s), cost 1.6× less, and passed 5 of 5 to its 5.

  • Against Claude Sonnet 5.5: Kimi K3 answered 3.2× later (5.4 s to 1.7 s), cost 1.5× more, and passed 5 of 5 to its 5.

  • Against Claude Sonnet 5: Kimi K3 answered 1.4× later (5.4 s to 3.9 s), cost 1.1× more, and passed 5 of 5 to its 5.

  • Against Claude Haiku 5.5: Kimi K3 answered 3.1× later (5.4 s to 1.7 s), cost 25.5× more, and passed 5 of 5 to its 5.

  • Against Claude Haiku 4.5: Kimi K3 answered 2.4× later (5.4 s to 2.2 s), cost 2.4× more, and passed 5 of 5 to its 5.

  • Against GPT-6 Astra: Kimi K3 answered 2.1× later (5.4 s to 2.6 s), cost 2.3× less, and passed 5 of 5 to its 5.

  • Against GPT-6.1 Sol: Kimi K3 answered 3.0× later (5.4 s to 1.8 s), cost 4.3× more, and passed 5 of 5 to its 5.

  • Against GPT-6 Sol: Kimi K3 answered 2.5× later (5.4 s to 2.2 s), cost 2.1× more, and passed 5 of 5 to its 5.

  • Against GPT-6 Luna: Kimi K3 answered 2.5× later (5.4 s to 2.2 s), cost 38.0× more, and passed 5 of 5 to its 4. Only Kimi K3 passed Pens at 3 for $4.

  • Against Gemini 4 Argon: Kimi K3 answered 1.2× later (5.4 s to 4.4 s), cost 2.4× more, and passed 5 of 5 to its 5.

  • Against Gemini 3.1 Pro (preview): Kimi K3 answered 1.5× sooner (5.4 s to 8.3 s), cost 2.7× less, and passed 5 of 5 to its 5.

  • Against Gemini 3.8 Flash: Kimi K3 answered 1.1× later (5.4 s to 4.7 s), cost 2.6× more, and passed 5 of 5 to its 5.

  • Against DeepSeek V4.1 Flash: Kimi K3 answered 5.8× later (5.4 s to 0.9 s), cost 15.0× more, and passed 5 of 5 to its 5.

  • Against DeepSeek V4 Pro: Kimi K3 answered 2.0× later (5.4 s to 2.7 s), cost 4.3× more, and passed 5 of 5 to its 5.

  • Against Grok 4.7: Kimi K3 answered 2.2× sooner (5.4 s to 11.7 s), cost 1.9× less, and passed 5 of 5 to its 5.

  • Against GLM 5.3: Kimi K3 answered 5.7× later (5.4 s to 0.9 s), cost 10.6× more, and passed 5 of 5 to its 5.

  • Against GLM 5.3 Flash: Kimi K3 answered 2.0× later (5.4 s to 2.7 s), cost 24.9× more, and passed 5 of 5 to its 5.

  • Against Mistral Large 4: Kimi K3 answered within 10% of its time (5.4 s to 5.1 s), cost 2.0× more, and passed 5 of 5 to its 5.

How these runs were done

Final answer. Automatic. The reply's last “Final answer:” line must hold the right value.

How the runs were done, and every math prompt.

Kimi K3, and math, elsewhere

Questions

Is Kimi K3 good for math?

In our test runs it passed 5 of 5 math prompts, tied 1st of the 20 models in llmwise. Every reply is on this page, so you can judge them yourself.

How many of my messages does a math reply from Kimi K3 use?

1 message each on Pro, where it has 125 a month on Pro. The price of a message is fixed and shown before you send it, however long the reply.

How were these runs done?

The same way for every model: each prompt sent through llmwise's own pipeline, each reply checked the same way. The methods page has every prompt and how each is scored.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.

Start freeSee pricing