Skip to content

Free tool · Math

AI math solver

Type any math problem, from a percentage to an integral, and get a numbered, checked solution: 1 free run a day on Claude Haiku 5.5, no account needed. Signed in, pick any model and see every step.

1 free run a day, no account needed

Claude Haiku 5.5

Up to 1,000 characters. No files, tools or web search in a free run. One free run a browser, and one a network each day, shared with the free message on our home page.

Try an example:

By running it, you agree to the Terms and Privacy Policy and confirm you’re 18 or older. We keep what you send and its answer for 7 days, or in your account if you sign up.

Test runs checked . Updated .

AI math solver: three examples

Three inputs and the solution a free run gave each, made once on September 28, 2026 with GPT-6 Luna and shown as it came back. A free run is on Claude Haiku 5.5 now. The example buttons in the box above show the same.

Solve a quadratic

Solve for x: 2x^2 - 7x + 3 = 0

1. Factor the quadratic by finding two numbers whose product is 2·3 = 6 and whose sum is -7: -6 and -1.

2x^2 - 7x + 3 = 2x^2 - 6x - x + 3

2. Group the terms and factor each group.

= 2x(x - 3) - 1(x - 3)

= (2x - 1)(x - 3)

3. Set each factor equal to zero, since their product is zero.

2x - 1 = 0, so x = 1/2

x - 3 = 0, so x = 3

4. Check by substitution.

For x = 1/2: 2(1/2)^2 - 7(1/2) + 3 = 1/2 - 7/2 + 3 = 0.

For x = 3: 2(3)^2 - 7(3) + 3 = 18 - 21 + 3 = 0.

Final answer: x = 1/2 or x = 3
A percentage word problem

A phone costs ₹18,000 after a 25% discount. What was its price before the discount?

1. A 25% discount means the phone costs 100% - 25% = 75% of its original price.

2. Let the original price be P.
   0.75P = 18,000

3. Divide both sides by 0.75.
   P = 18,000 / 0.75 = 24,000

4. Check: 25% of ₹24,000 is ₹6,000, and ₹24,000 - ₹6,000 = ₹18,000.

Final answer: ₹24,000
A derivative

Find the derivative of f(x) = x^3 · ln(x), and the slope of the curve at x = 1.

1. The function is f(x) = x^3 · ln(x). Since it contains ln(x), we consider x > 0.

2. Use the product rule: the derivative of u · v is u' · v + u · v'. Here, u = x^3 and v = ln(x).

3. Differentiate each factor:
u' = 3x^2
v' = 1/x

4. Apply the product rule:
f'(x) = 3x^2 · ln(x) + x^3 · (1/x)

5. Simplify:
f'(x) = 3x^2 ln(x) + x^2

6. Find the slope at x = 1 by substituting 1:
f'(1) = 3(1)^2 ln(1) + (1)^2

7. Since ln(1) = 0:
f'(1) = 0 + 1 = 1

8. Check: substituting x = 1 into the derivative gives 1, so the slope there is 1.

Final answer: f'(x) = 3x^2 ln(x) + x^2; the slope at x = 1 is 1.

Which AI model is best at this? We tested every one

Every one of the 20 models we ran passed all 5 of our AI math solver test inputs, so choose on speed and price: DeepSeek V4.1 Flash was quickest, at 1.1 s a run on average. Claude Haiku 5.5, the model a free run is on, took 3.0 s.

Run on October 15, 2026: 5 inputs on 20 models, each with this tool's own instructions.

Each model's results on the tool's test inputs
ModelPassedHard onesMessages used on ProCost per runTime per run
GPT-6 LunaOpenAI5 of 52 of 21 each, of 60 a day on Pro$0.00012.8 s
GLM 5.3 FlashZ.ai5 of 52 of 21 each, of 60 a day on Pro$0.00022.2 s
Claude Haiku 5.5AnthropicThe free run's model5 of 52 of 21 each, of 60 a day on Pro$0.00033.0 s
DeepSeek V4.1 FlashDeepSeek5 of 52 of 21 each, of 60 a day on Pro$0.00051.1 s
Gemini 3.8 FlashGoogle5 of 52 of 21 each, of 250 a month on Pro$0.00102.6 s
GLM 5.3Z.ai5 of 52 of 21 each, of 250 a month on Pro$0.00111.2 s
DeepSeek V4 ProDeepSeek5 of 52 of 21 each, of 250 a month on Pro$0.00223.4 s
Claude Haiku 4.5Anthropic5 of 52 of 21 each, of 250 a month on Pro$0.00243.5 s
Mistral Large 4Mistral5 of 52 of 21 each, of 250 a month on Pro$0.004725.7 s
Grok 4.7xAI5 of 52 of 21 each, of 250 a month on Pro$0.007010.8 s
GPT-6.1 SolOpenAI5 of 52 of 21 each, of 125 a month on Pro$0.00153.2 s
GPT-6 SolOpenAI5 of 52 of 21 each, of 125 a month on Pro$0.00163.8 s
Kimi K3Moonshot5 of 52 of 21 each, of 125 a month on Pro$0.004325.9 s
Claude Sonnet 5Anthropic5 of 52 of 21 each, of 125 a month on Pro$0.00444.4 s
Claude Sonnet 5.5Anthropic5 of 52 of 21 each, of 125 a month on Pro$0.00513.2 s
Gemini 3.1 Pro (preview)Google5 of 52 of 21 each, of 125 a month on Pro$0.007810.8 s
Gemini 4 ArgonGoogle5 of 52 of 21 each, of 62 a month on Pro$0.003513.3 s
Claude Opus 5.5Anthropic5 of 52 of 21 each, of 62 a month on Pro$0.01165.1 s
GPT-6 AstraOpenAI5 of 52 of 21 each, of 31 a month on Pro$0.00693.1 s
Claude Fable 5.1Anthropic5 of 52 of 21 each, of 31 a month on Pro$0.02425.4 s
Passed: of the 5 test inputs, how many answers passed their check (an answer the provider failed to give isn't counted). Messages used: what each run counted as on Pro, and Pro's count on the model. Cost: what OpenRouter charged us per run, on average; in llmwise you pay per message, not per token. Time: from sending to the whole answer.

Side by side: a remainder of a large power

The same input on GPT-6 Luna and Claude Haiku 5.5, as each answered it, unedited.

Solve this math problem step by step. Number the steps, check the result, and end with a line that starts "Final answer:".

---
What is the remainder when 7^100 is divided by 13?
GPT-6 Luna
1. Reduce powers of 7 modulo 13:
   7^2 = 49 ≡ 10 (mod 13).
   7^3 ≡ 10 × 7 = 70 ≡ 5 (mod 13).
   7^6 ≡ 5^2 = 25 ≡ 12 ≡ -1 (mod 13).

2. Therefore, 7^12 ≡ (-1)^2 = 1 (mod 13).

3. Write 100 = 8 × 12 + 4, so:
   7^100 ≡ (7^12)^8 × 7^4 ≡ 7^4 (mod 13).

4. Compute the remaining power:
   7^4 ≡ (7^2)^2 ≡ 10^2 = 100 ≡ 9 (mod 13).

5. Check: 100 = 13 × 7 + 9, so the remainder is indeed 9.

Final answer: 9
Claude Haiku 5.5
Step 1: Find the pattern of powers of 7 modulo 13.
7^1 = 7 (mod 13), so 7^1 ≡ 7.

Step 2: Compute the next powers.
7^2 = 49. Since 49 = 3·13 + 10, 7^2 ≡ 10 (mod 13).

Step 3: Continue.
7^3 ≡ 10·7 = 70 = 5·13 + 5, so 7^3 ≡ 5 (mod 13).

Step 4: Continue.
7^4 ≡ 5·7 = 35 = 2·13 + 9, so 7^4 ≡ 9 (mod 13).

Step 5: Continue.
7^5 ≡ 9·7 = 63 = 4·13 + 11, so 7^5 ≡ 11 (mod 13).

Step 6: Continue.
7^6 ≡ 11·7 = 77 = 5·13 + 12, so 7^6 ≡ 12 (mod 13).

Step 7: Notice 12 ≡ -1 (mod 13), so 7^6 ≡ -1 and 7^12 ≡ (-1)^2 = 1 (mod 13).
The order of 7 modulo 13 divides 12, and we have 7^6 ≡ -1 ≠ 1, so the order is 12 (we also checked the smaller candidates 1, 2, 3, 4, 6 are not valid since 7^6 ≠ 1 and 7^4 ≠ 1, 7^3 ≠ 1, 7^2 ≠ 1, and the order must divide 12 and cannot be 6, so it is 4 or 12 or 3 or 2 or 1; 7^4 ≡ 9 ≠ 1 rules out 1, 2, 4, and 7^3 ≡ 5 ≠ 1 rules out 3, so the order is 12).

Step 8: Reduce the exponent.
100 = 8·12 + 4, so 7^100 = (7^12)^8 · 7^4 ≡ 1^8 · 7^4 ≡ 7^4 (mod 13).

Step 9: Use the value from Step 4.
7^4 ≡ 9 (mod 13).

Step 10: Check.
Direct check by another route: 7^4 = 2401. Dividing 2401 by 13: 13·184 = 2392, and 2401 − 2392 = 9. This matches 9.
Also check 7^100 ≡ 7^4 via Fermat's little theorem: 7^12 ≡ 1 (mod 13), and 100 mod 12 = 4, which agrees with Step 8.

Final answer: 9

How the inputs are scored

  • Final answer: the answer's last “Final answer:” line must hold the right value; nothing else in it is judged.

The inputs, each linked to every model's reply where it has a page of its own: a linear equation, a discount word problem, two dice that add up to 8, a remainder of a large power, and a definite integral by parts. Every model gets the tool's own instructions and up to Pro's reply limit. The method is the same as our other test runs.

How to use the AI math solver

  1. Type or paste one problem, up to 1,000 characters. Write powers as x^2, roots as sqrt(x) and fractions as (a+b)/c.

  2. Pick how much explaining you want, and the language.

  3. Press “Solve it”. The solution streams in; without an account you see its first part (1 free run a day, no account needed).

  4. Signed in, pick any model your plan has, get the whole solution, and carry on in a chat to ask about any step.

Tips for a better answer

  • One problem at a time

    One problem a run. A worksheet of ten gets a rushed answer to each; ten runs (or one chat, signed in) get ten careful ones.

  • Give it everything the problem gives

    Copy the problem exactly, with every number and unit. For a word problem, keep the whole story: the numbers often hide in it.

  • Name the method you need

    Add “by factoring”, “without a calculator” or “using the chain rule” to the problem, and the solution follows your class's method.

  • Use it to practise, not just to check

    Cover the solution after step 1 and try the next step yourself, then compare. It's the fastest way to learn the method rather than the answer.

What a good solution looks like

  • Every step follows from the last

    Each line should follow from the one above it by one move you can name: expand, collect like terms, divide both sides. A jump you can't follow is where mistakes hide.

  • The answer is checked

    A good solution puts the answer back into the original problem (or estimates it) before it states it. Do the same: it takes a few seconds and catches most slips.

  • Watch for the usual slips

    Moving a term across the equals sign without changing its sign, taking only the positive square root, and rounding in the middle of a calculation are the three slips we see most.

  • Units carried through

    Word problems need their units all the way through: if the question is in km/h and minutes, the answer should say which it's in.

What's free, and what signing up adds

  • Without an account: 1 free run a day, on Claude Haiku 5.5, shared with the free message on our home page. It takes up to 1,000 characters, writes up to about 600 tokens (roughly 450 words), and shows you the start of the solution.

  • Signing up is free, with an email link and no card. llmwise accounts are for people 18 and older. The solution you ran comes with you as your first chat, and you get 5 free messages on Claude Haiku 5.5.

  • Signed in, the tool runs on the model you pick, on your plan, shows the whole answer, and you can carry on in the chat with any model, or ask another model the same thing.

More free tools are on the tools page, and what each plan gives is on pricing.

Questions

Is the AI math solver free?

Yes. Without an account you get 1 run a day on Claude Haiku 5.5, shared with the free message on our home page: you see the start of the solution, and the rest when you sign up. Signing up is free and adds 5 free messages on Claude Haiku 5.5, with no card. Every other model in llmwise is on a paid plan.

What kinds of math can it solve?

Arithmetic, percentages, algebra, equations and inequalities, geometry and trigonometry, probability and statistics, and calculus (limits, derivatives, integrals), including word problems. Long proofs and problems with many parts may not fit in one free run's answer; ask one part at a time.

Can it solve a math problem from a photo?

Not in the free run, which takes typed text. After you sign up you can attach a photo of the problem in a chat, pick a model that reads images, and ask for the same step-by-step solution.

How do I know the answer is right?

It shows every step and the check it did, and ends with a line that starts “Final answer:”. Put that answer back into the original problem yourself: if both sides agree, it's right.

Can it explain the steps, not just give the answer?

Yes: pick “Every step, explained” and it says why each step follows, in plain words. You can also ask for the method your class uses, such as “by factoring” or “using the quadratic formula”, in the problem box.

Which AI is best for this, in your tests?

Every one of the 20 models we ran passed all 5 of our AI math solver test inputs, so choose on speed and price: DeepSeek V4.1 Flash was quickest, at 1.1 s a run on average. Claude Haiku 5.5, the model a free run is on, took 3.0 s. The table on this page has every model's result, what a run counts as on Pro, and the time it took.

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.