Mistral Large 4 · Math
Mistral Large 4 for math: our test runs
Mistral Large 4 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.
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 October 9, 2026, Mistral Large 4 passed 5 of 5 of our math prompts (both hard ones): tied 1st of 19 models. Each reply counted as 1 message on Pro, where it has 250 a month, and cost $0.0018 to run on average.
Prompt by prompt
Each of the 5 math prompts, and how Mistral Large 4's reply did.
| Prompt | Result | Cost | Time |
|---|---|---|---|
| A discount, then sales taxEveryday | Passed: Final answer 66: right. | $0.0009 | 3.5 s |
| Pens at 3 for $4Everyday | Passed: Final answer Buy three 3-packs plus one single pen, for 13.50 total.: right. | $0.0022 | 22.0 s |
| Compound interest over three yearsEveryday | Passed: Final answer 315.25: right. | $0.0014 | 5.1 s |
| Four-digit numbers whose digits sum to 9Hard | Passed: Final answer 165: right. | $0.0007 | 2.6 s |
| The highest of three dice is a 5Hard | Passed: Final answer 61/216: right. | $0.0039 | 17.9 s |
Every math reply from Mistral Large 4
A discount, then sales tax
An everyday prompt: “A discount, then sales tax”, in full.
Passed: Final answer 66: right.
Discount: 25% of $80 = $20, so discounted price is $80 − $20 = $60. Tax: 10% of $60 = $6, so final cost is $60 + $6 = $66. Final answer: $66
406 tokens in, 305 out (210 of them reasoning) · 3.5 s · $0.0009 · 1 message on Pro · answered by mistralai/mistral-large-4-0 via Mistral ·
Pens at 3 for $4
An everyday prompt: “Pens at 3 for $4”, in full.
Passed: Final answer Buy three 3-packs plus one single pen, for 13.50 total.: right.
- 3-pack deal: $4 for 3 pens ≈ $1.33/pen — cheaper than the $1.50 single. - For exactly 10 pens, use as many 3-packs as possible without going over: 3 packs = 9 pens for 3 × $4 = $12. - Remaining 1 pen bought singly: $1.50. - Total: $12 + $1.50 = $13.50. Final answer: Buy three 3-packs plus one single pen, for $13.50 total.
408 tokens in, 917 out (616 of them reasoning) · 22.0 s · $0.0022 · 1 message on Pro · answered by mistralai/mistral-large-4-0 via Mistral ·
Compound interest over three years
An everyday prompt: “Compound interest over three years”, in full.
Passed: Final answer 315.25: right.
Using compound interest formula: $A = P(1 + r)^t$ $A = 2000(1.05)^3 = 2000 \times 1.157625 = \$2,315.25$ Interest earned = $\$2,315.25 - \$2,000 = \$315.25$ (Year-by-year: Year 1 earns \$100.00, Year 2 earns \$105.00, Year 3 earns \$110.25) Final answer: $315.25
408 tokens in, 558 out (304 of them reasoning) · 5.1 s · $0.0014 · 1 message on Pro · answered by mistralai/mistral-large-4-0 via Mistral ·
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 four-digit number be \(abcd\), where \(a \ge 1\) and \(b,c,d \ge 0\). We need \[ a+b+c+d=9. \] Set \(a'=a-1\), so \(a',b,c,d \ge 0\). Then \[ a'+b+c+d=8. \] By stars and bars, the number of nonnegative solutions is \[ \binom{8+4-1}{4-1}=\binom{11}{3}=165. \] Final answer: 165383 tokens in, 219 out (72 of them reasoning) · 2.6 s · $0.0007 · 1 message on Pro · answered by mistralai/mistral-large-4-0 via Mistral ·
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.
Total outcomes when rolling three dice: $6^3 = 216$. For the highest number to be exactly 5: - No die can show 6, so all dice show values from $\{1,2,3,4,5\}$: $5^3 = 125$ outcomes. - But we must exclude cases where the highest is 4 or less (i.e., all dice from $\{1,2,3,4\}$): $4^3 = 64$ outcomes. Favorable outcomes: $125 - 64 = 61$. Probability: $\frac{61}{216}$ (61 is prime and doesn't divide 216, so this is in lowest terms). Final answer: $\frac{61}{216}$399 tokens in, 1,723 out (1,226 of them reasoning) · 17.9 s · $0.0039 · 1 message on Pro · answered by mistralai/mistral-large-4-0 via Mistral ·
How it compares on math
Mistral Large 4 beside the job's picks among every model. 18 models share the hard-problems pick, Mistral Large 4 among them.
| Model | Passed | On Pro | Cost per reply |
|---|---|---|---|
| GLM 5.3Our pick: hard problems (shared), best value | 5 of 5 | 250 a month on Pro | $0.0004 |
| Claude Haiku 5.5Our pick: hard problems (shared), everyday | 5 of 5 | 60 a day on Pro | $0.0001 |
| Mistral Large 4Our pick: hard problems (shared) | 5 of 5 | 250 a month on Pro | $0.0018 |
Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: 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.
Mistral Large 4 next to each model on math
Mistral Large 4 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: Mistral Large 4 answered within 10% of its time (5.1 s to 4.6 s), cost 6.4× less, and passed 5 of 5 to its 5.
Against Claude Opus 5.5: Mistral Large 4 answered 1.3× later (5.1 s to 3.9 s), cost 3.3× less, and passed 5 of 5 to its 5.
Against Claude Sonnet 5.5: Mistral Large 4 answered 3.1× later (5.1 s to 1.7 s), cost 1.4× less, and passed 5 of 5 to its 5.
Against Claude Sonnet 5: Mistral Large 4 answered 1.3× later (5.1 s to 3.9 s), cost 1.8× less, and passed 5 of 5 to its 5.
Against Claude Haiku 5.5: Mistral Large 4 answered 2.9× later (5.1 s to 1.7 s), cost 12.4× more, and passed 5 of 5 to its 5.
Against Claude Haiku 4.5: Mistral Large 4 answered 2.3× later (5.1 s to 2.2 s), cost 1.2× more, and passed 5 of 5 to its 5.
Against GPT-6 Astra: Mistral Large 4 answered 2.0× later (5.1 s to 2.6 s), cost 4.7× less, and passed 5 of 5 to its 5.
Against GPT-6.1 Sol: Mistral Large 4 answered 2.9× later (5.1 s to 1.8 s), cost 2.1× more, and passed 5 of 5 to its 5.
Against GPT-6 Sol: Mistral Large 4 answered 2.3× later (5.1 s to 2.2 s), cost within 10% of it, and passed 5 of 5 to its 5.
Against GPT-6 Luna: Mistral Large 4 answered 2.3× later (5.1 s to 2.2 s), cost 18.6× more, and passed 5 of 5 to its 4. Only Mistral Large 4 passed Pens at 3 for $4.
Against Gemini 3.1 Pro (preview): Mistral Large 4 answered 1.6× sooner (5.1 s to 8.3 s), cost 5.5× less, and passed 5 of 5 to its 5.
Against Gemini 3.8 Flash: Mistral Large 4 answered within 10% of its time (5.1 s to 4.7 s), cost 1.3× more, and passed 5 of 5 to its 5.
Against DeepSeek V4.1 Flash: Mistral Large 4 answered 5.5× later (5.1 s to 0.9 s), cost 7.3× more, and passed 5 of 5 to its 5.
Against DeepSeek V4 Pro: Mistral Large 4 answered 1.9× later (5.1 s to 2.7 s), cost 2.1× more, and passed 5 of 5 to its 5.
Against Grok 4.7: Mistral Large 4 answered 2.3× sooner (5.1 s to 11.7 s), cost 4.0× less, and passed 5 of 5 to its 5.
Against Kimi K3: Mistral Large 4 answered within 10% of its time (5.1 s to 5.4 s), cost 2.0× less, and passed 5 of 5 to its 5.
Against GLM 5.3: Mistral Large 4 answered 5.4× later (5.1 s to 0.9 s), cost 5.2× more, and passed 5 of 5 to its 5.
Against GLM 5.3 Flash: Mistral Large 4 answered 1.9× later (5.1 s to 2.7 s), cost 12.2× 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.
Mistral Large 4, and math, elsewhere
- Mistral Large 4: price, limits and messages on every plan
- The best AI for math: our picks among every model
- Mistral Large 4 for coding: our test runs
- Mistral Large 4 for writing: our test runs
- Mistral Large 4 for data analysis: our test runs
- Analyze a CSV with code: the prompt, and its price
- AI prompts for students, tested on four models
- AI prompts for teachers, tested with every reply
- Our test runs: 50 prompts on every model
Questions
Is Mistral Large 4 good for math?
In our test runs it passed 5 of 5 math prompts, tied 1st of the 19 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 Mistral Large 4 use?
1 message each on Pro, where it has 250 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.