Advanced Placement
Skip to the guide

AP Precalculus

2025 AP Precalculus free-response questions, solved

Updated · For the May 2027 AP exam

The 2025 free-response section had four 6-point questions: a decreasing function given by a table alongside a cubic, a quadratic model of a song’s total plays, a vibrating guitar string modeled by a sinusoid, and logarithmic, trigonometric and exponential equations to solve by hand. The answers are in the first table, followed by our worked solutions and explanations. Keep the official question booklet open alongside them, and use the official scoring guidelines to mark your work.

The answers at a glance

2025 AP Precalculus free-response answers
PartAnswerStudents who earned it
Question 1Function Conceptsmean 3.51 of 6
A(i)h(1)=g(1.75)=0.333h(1) = g(1.75) = 0.33364%
A(ii)f−1(3.5)=0f^{-1}(3.5) = 072%
B(i)x=−1.233x = -1.233, x=1.578x = 1.578, x=5.643x = 5.64356%
B(ii)lim⁡x→∞g(x)=−∞\lim_{x \to \infty} g(x) = -\infty56%
C(i)Exponential73%
C(ii)Over input steps of 1, each output is 0.5 times the one before30%
Question 2Modeling a Non-Periodic Context: song playsmean 2.16 of 6
A(i)c=25c = 25, 4a+2b+c=304a + 2b + c = 30, 16a+4b+c=3416a + 4b + c = 3453%
A(ii)a=−0.125a = -0.125, b=2.75b = 2.75, c=25c = 2548%
B(i)2.25 thousand plays per month63%
B(ii)28.375 thousand plays41%
B(iii)DD is concave down, so the secant line lies below its graph6%
CTotal plays never decrease, but DD decreases after its maximum at t=11t = 11, so the domain ends at t=11t = 115%
Question 3Modeling a Periodic Context: the guitar stringmean 3.13 of 6
A, heightsF 2, G 0, J −2, K 0, P 277%
A, times00, 1800\frac{1}{800}, 2800\frac{2}{800}, 3800\frac{3}{800}, 4800\frac{4}{800}34%
B, aa and dda=2a = 2, d=0d = 067%
B, bb and ccb=400πb = 400\pi, c=1800c = \frac{1}{800}18%
C(i)d: negative and decreasing67%
C(ii)Concave up; the rate of change of hh is increasing51%
Question 4Symbolic Manipulationsmean 1.93 of 6
A(i)x=9x = 953%
A(ii)x=π6x = \frac{\pi}{6}43%
B(i)log⁡2(8x)\log_2(8x)45%
B(ii)6tan⁡x6\tan x28%
C, first point(ex−4)(ex+3)=0(e^x - 4)(e^x + 3) = 014%
C, second pointx=ln⁡4x = \ln 410%
Percentages are the mean score on each 1-point item from the 2025 Chief Reader report; question means are from the 2025 scoring statistics. Other correct forms are accepted where the guidelines allow them; see each solution. Some first-point awards include partial credit for work elsewhere in the part; the percentage is the recorded point award, not always success on the named task alone.

Question 1: Function Concepts

Graphing calculator required. Question 1 is on page 3 of the PDF.

Question 1, worked solution

A (i) f(1)=1.75f(1) = 1.75, so h(1)=g(1.75)=−0.167(1.75)3+(1.75)2−1.834=0.333h(1) = g(1.75) = -0.167(1.75)^3 + (1.75)^2 - 1.834 = 0.333.

A (ii) f(0)=3.5f(0) = 3.5, so f−1(3.5)=0f^{-1}(3.5) = 0.

B (i) x=−1.233x = -1.233, x=1.578x = 1.578 and x=5.643x = 5.643

B (ii) lim⁡x→∞g(x)=−∞\lim_{x \to \infty} g(x) = -\infty

C (i) Exponential.

C (ii) The inputs in the table go up by 1 each step, and each output is 0.5 times the one before: 714=3.57=1.753.5=0.8751.75=0.5\frac{7}{14} = \frac{3.5}{7} = \frac{1.75}{3.5} = \frac{0.875}{1.75} = 0.5. Over equal-length input intervals the outputs change by the same factor, so ff is best modeled by an exponential function.

Why it works

  • The composition uses 1.75 as the input to the cubic; it does not use 1 again.
  • An inverse lookup runs backward through the input-output pairs.
  • A graphing window must include the zero near 5.6 as well as the other two. The table’s repeated ratio explains the exponential model independently of any regression.

On the calculator. For A(i), store 1.75 or type it in directly. For B(i), graph y=−0.167x3+x2−1.834y = -0.167x^3 + x^2 - 1.834 and find each xx-intercept: tap them in Desmos, or use 2nd, TRACE (CALC), 2: zero on a TI-84, once per zero. Make sure the window shows all three: the third is near x=5.6x = 5.6, and the report lists missing solutions as a common error.

Question 2: Modeling a Non-Periodic Context

Graphing calculator required. Question 2 is on page 4 of the PDF.

Question 2, worked solution

A (i) D(0)=25D(0) = 25, D(2)=30D(2) = 30 and D(4)=34D(4) = 34 give c=25c = 25, 4a+2b+c=304a + 2b + c = 30 and 16a+4b+c=3416a + 4b + c = 34.

A (ii) With c=25c = 25: 4a+2b=54a + 2b = 5 and 16a+4b=916a + 4b = 9. Doubling the first and subtracting it from the second gives 8a=−18a = -1, so a=−0.125a = -0.125, then b=2.75b = 2.75 and c=25c = 25.

B (i) D(4)−D(0)4−0=34−254=2.25\frac{D(4) - D(0)}{4 - 0} = \frac{34 - 25}{4} = 2.25 thousand plays per month

B (ii) 25+2.25(1.5)=28.37525 + 2.25(1.5) = 28.375 thousand plays

B (iii) AtA_t is the yy-coordinate of a point on the secant line through (0,D(0))(0, D(0)) and (4,D(4))(4, D(4)). Since a<0a < 0, the graph of DD is concave down on 0<t<40 < t < 4, so the secant line lies below the graph of DD there. So At<D(t)A_t < D(t) for 0<t<40 < t < 4.

C The total number of plays since release can never decrease. DD has its absolute maximum at t=−b2a=11t = -\frac{b}{2a} = 11 and decreases after that, so DD can only model the total plays up to t=11t = 11: the right end of its domain is t=11t = 11.

Why it works

  • Subtracting twice the first reduced equation from the second isolates aa; substituting back gives bb.
  • The secant line joins the interval’s endpoints. The middle measurement is above that line, so starting the estimate there changes the model.
  • A cumulative play count cannot decrease. That contextual fact is what makes the quadratic’s maximum a useful boundary.

On the calculator. For A(ii), put the three points in a table and run a quadratic regression (y1∼ax12+bx1+cy_1 \sim ax_1^2 + bx_1 + c in Desmos, or QuadReg on a TI-84). Three points with distinct inputs fit a polynomial of degree at most two; these three give a quadratic. You still need the three equations on paper for A(i): the regression output doesn’t earn that point.

Why B(iii) and C went to so few students. They need no new math. B(iii) needs concavity and the secant line in the same sentence; C needs what the model does and what the real total can do, side by side. The FRQ guide has examples of answers that do and don’t earn these points.

Question 3: Modeling a Periodic Context

No calculator. Question 3 is on page 6 of the PDF.

Question 3, worked solution

A 200 cycles a second means one cycle takes 1200\frac{1}{200} second, so the labeled points are 1800\frac{1}{800} second apart. X starts at its highest point, so F is at t=0t = 0. F (0,2)(0, 2), G (1800,0)\left(\frac{1}{800}, 0\right), J (2800,−2)\left(\frac{2}{800}, -2\right), K (3800,0)\left(\frac{3}{800}, 0\right), P (4800,2)\left(\frac{4}{800}, 2\right).

B a=2a = 2, d=0d = 0, b=2π1/200=400πb = \frac{2\pi}{1/200} = 400\pi. The first peak is at t=0t = 0, so c=14⋅1200−0=1800c = \frac{1}{4} \cdot \frac{1}{200} - 0 = \frac{1}{800}. That gives h(t)=2sin⁡(400π(t+1800))h(t) = 2\sin\left(400\pi\left(t + \frac{1}{800}\right)\right). Check: h(0)=2sin⁡π2=2h(0) = 2\sin\frac{\pi}{2} = 2.

C (i) d: hh is negative and decreasing.

C (ii) The graph of hh is concave up on (t1,t2)(t_1, t_2), and the rate of change of hh is increasing.

Why it works

  • Frequency counts cycles per second; period measures seconds per cycle. Divide the period by four to space the five marked times.
  • In b(t+c)b(t+c), cc has units of time. The angle shift in bt+bcbt+bc is bcbc, not cc.
  • From G to J, hh decreases toward its minimum while its slope increases toward zero. “The rate of change of hh is increasing” avoids ambiguity.

**The cc trap.** Because cc sits inside b(t+c)b(t + c), it’s measured in seconds, and here it’s tiny: 1800\frac{1}{800}. A value like π2\frac{\pi}{2} belongs to the other form, sin⁡(bt+c)\sin(bt + c). The cosine model h(t)=2cos⁡(400πt)h(t) = 2\cos(400\pi t) is also true but isn’t the form asked for; the guidelines gave it 1 of the 2 points in part B.

Question 4: Symbolic Manipulations

No calculator. Question 4 is on page 7 of the PDF.

Question 4, worked solution

A (i) 2log⁡3x=42\log_3 x = 4, so log⁡3x=2\log_3 x = 2 and x=32=9x = 3^2 = 9.

A (ii) 4cos⁡2x=34\cos^2 x = 3, so cos⁡2x=34\cos^2 x = \frac{3}{4} and cos⁡x=±32\cos x = \pm\frac{\sqrt{3}}{2}. In [0,π2)\left[0, \frac{\pi}{2}\right), cosine is positive, so x=π6x = \frac{\pi}{6}.

B (i) 3log⁡22=log⁡223=log⁡283\log_2 2 = \log_2 2^3 = \log_2 8, so j(x)=log⁡2x+log⁡28=log⁡2(8x)j(x) = \log_2 x + \log_2 8 = \log_2(8x).

B (ii) csc⁡2x−1=cot⁡2x\csc^2 x - 1 = \cot^2 x, so k(x)=6tan⁡xcot⁡2x=6cot⁡x=6tan⁡xk(x) = \frac{6}{\tan x\cot^2 x} = \frac{6}{\cot x} = 6\tan x.

C e2x−ex−12=(ex)2−ex−12=(ex−4)(ex+3)=0e^{2x} - e^x - 12 = (e^x)^2 - e^x - 12 = (e^x - 4)(e^x + 3) = 0. ex=−3e^x = -3 is impossible because ex>0e^x > 0, so ex=4e^x = 4 and x=ln⁡4x = \ln 4.

Why it works

  • Only the first-quadrant cosine solution lies in the requested interval.
  • Use exe^x as one variable before factoring; the negative root for that variable cannot produce a real xx.
  • The logarithms require x>0x > 0. The tangent rewrite retains sin⁡x≠0\sin x \ne 0 and cos⁡x≠0\cos x \ne 0 from the original expression.

Why part C stopped so many students. e2xe^{2x} is (ex)2(e^x)^2, and once you see that, the equation is an ordinary quadratic. The Chief Reader’s note was that few responses recognized it. If an equation has e2xe^{2x} and exe^x, or sin⁡2x\sin^2 x and sin⁡x\sin x, try the substitution first.

From points to a score

Mark your work with the official 2025 scoring guidelines, including their partial-credit and decimal-presentation rules. Record each question’s score below, then enter those scores in the score calculator. The calculator uses the May 2027 multiple-choice format, so do not enter a 40-question 2025 multiple-choice total as though it were out of 42.

Record your 2025 free-response scores

How each question type is scored, and how to write the explanations that so few students earned, is in the AP Precalculus FRQ guide. More practice: the 2026 solutions and the 2024 solutions.

Sources

  1. 2025 AP Precalculus Free-Response Questions (PDF)
  2. 2025 AP Precalculus Scoring Guidelines (PDF)
  3. 2025 AP Precalculus Chief Reader Report: per-point means and common errors (PDF)
  4. 2025 AP Precalculus Scoring Statistics (PDF)
  5. AP Precalculus score distributions (AP Students)
  6. AP Precalculus past exam questions, 2024–2026 (AP Central)
  7. Texas Instruments: finding intersections on TI-83/84 calculators
  8. Texas Instruments: TI-84 Plus CE best practices (PDF), CALC menu and zero/intersection tools