diff --git a/tests/704_math.py b/tests/704_math.py
index 15688789..8d68341d 100644
--- a/tests/704_math.py
+++ b/tests/704_math.py
@@ -1,64 +1,259 @@
-from math import log, log10, log2, sin, cos, tan, e, pi, isnan, isinf, fabs, floor, ceil, sqrt
+import math
+
+# ============================================================
+# Helper function for floating point comparison
+# ============================================================
 
 def isclose(a, b):
     return abs(a-b) < 0.000001
 
-assert isclose(e, 2.718281828459045), e
-assert isclose(pi, 3.141592653589793), pi
-assert isclose(log(10), 2.302585092994046), log(10)
-assert isclose(log10(10), 1.0), log10(10)
-assert isclose(log2(10), 3.321928094887362), log2(10)
-assert isclose(sin(0), 0.0), sin(0)
-assert isclose(cos(0), 1.0), cos(0)
-assert isclose(tan(0), 0.0), tan(0)
+# ============================================================
+# Constants: pi, e, inf, nan
+# ============================================================
 
-a = -0.1
-a = a**a
-assert isnan(a)
-assert not isinf(a)
-assert isinf(float("inf"))
+assert isclose(math.pi, 3.141592653589793)
+assert isclose(math.e, 2.718281828459045)
+assert math.isinf(math.inf)
+assert math.isnan(math.nan)
 
-assert isclose(fabs(-1.2), 1.2)
+# ============================================================
+# Rounding functions: ceil, floor, trunc, fabs
+# ============================================================
 
-assert floor(1.2) == 1
-assert floor(-1.2) == -2
-assert ceil(1.2) == 2
-assert ceil(-1.2) == -1
+# ceil - round up to nearest integer
+assert math.ceil(1.2) == 2
+assert math.ceil(-1.2) == -1
+assert math.ceil(2.0) == 2
+assert math.ceil(0) == 0
 
-assert isclose(sqrt(4), 2.0)
+# floor - round down to nearest integer
+assert math.floor(1.2) == 1
+assert math.floor(-1.2) == -2
+assert math.floor(2.0) == 2
+assert math.floor(0) == 0
 
-import math
+# trunc - truncate towards zero
+assert math.trunc(1.7) == 1
+assert math.trunc(-1.7) == -1
+assert math.trunc(2.0) == 2
+assert math.trunc(0) == 0
 
-# test fsum
+# fabs - absolute value (returns float)
+assert isclose(math.fabs(-1.2), 1.2)
+assert isclose(math.fabs(1.2), 1.2)
+assert isclose(math.fabs(0), 0.0)
+assert isclose(math.fabs(-0.0), 0.0)
+
+# ============================================================
+# Classification functions: isnan, isinf, isfinite, isclose
+# ============================================================
+
+# isnan - check if value is NaN
+assert math.isnan(float('nan'))
+assert math.isnan(math.nan)
+assert not math.isnan(1.0)
+assert not math.isnan(math.inf)
+
+# isinf - check if value is infinity
+assert math.isinf(float('inf'))
+assert math.isinf(float('-inf'))
+assert math.isinf(math.inf)
+assert not math.isinf(1.0)
+assert not math.isinf(math.nan)
+
+# isfinite - check if value is finite (not inf or nan)
+assert math.isfinite(1.0)
+assert math.isfinite(0.0)
+assert math.isfinite(-1.0)
+assert not math.isfinite(math.inf)
+assert not math.isfinite(-math.inf)
+assert not math.isfinite(math.nan)
+
+# isclose - check if two values are close
+assert math.isclose(1.0, 1.0)
+assert math.isclose(1.0, 1.0 + 1e-10)
+assert not math.isclose(1.0, 1.1)
+
+# ============================================================
+# Power and logarithmic functions: exp, log, log2, log10, pow, sqrt
+# ============================================================
+
+# exp - e raised to the power x
+assert isclose(math.exp(0), 1.0)
+assert isclose(math.exp(1), math.e)
+assert isclose(math.exp(2), math.e * math.e)
+
+# log - natural logarithm (base e) or logarithm with custom base
+assert isclose(math.log(1), 0.0)
+assert isclose(math.log(math.e), 1.0)
+assert isclose(math.log(10), 2.302585092994046)
+assert isclose(math.log(100, 10), 2.0)  # log base 10
+assert isclose(math.log(8, 2), 3.0)     # log base 2
+
+# log2 - logarithm base 2
+assert isclose(math.log2(1), 0.0)
+assert isclose(math.log2(2), 1.0)
+assert isclose(math.log2(8), 3.0)
+assert isclose(math.log2(10), 3.321928094887362)
+
+# log10 - logarithm base 10
+assert isclose(math.log10(1), 0.0)
+assert isclose(math.log10(10), 1.0)
+assert isclose(math.log10(100), 2.0)
+assert isclose(math.log10(1000), 3.0)
+
+# pow - x raised to the power y
+assert isclose(math.pow(2, 3), 8.0)
+assert isclose(math.pow(2, 0), 1.0)
+assert isclose(math.pow(2, -1), 0.5)
+assert isclose(math.pow(4, 0.5), 2.0)
+assert isclose(math.pow(0, 0), 1.0)
+
+# sqrt - square root
+assert isclose(math.sqrt(4), 2.0)
+assert isclose(math.sqrt(2), 1.4142135623730951)
+assert isclose(math.sqrt(0), 0.0)
+assert isclose(math.sqrt(1), 1.0)
+
+# ============================================================
+# Trigonometric functions: sin, cos, tan, asin, acos, atan, atan2
+# ============================================================
+
+# sin - sine
+assert isclose(math.sin(0), 0.0)
+assert isclose(math.sin(math.pi / 2), 1.0)
+assert isclose(math.sin(math.pi), 0.0)
+assert isclose(math.sin(-math.pi / 2), -1.0)
+
+# cos - cosine
+assert isclose(math.cos(0), 1.0)
+assert isclose(math.cos(math.pi / 2), 0.0)
+assert isclose(math.cos(math.pi), -1.0)
+assert isclose(math.cos(-math.pi), -1.0)
+
+# tan - tangent
+assert isclose(math.tan(0), 0.0)
+assert isclose(math.tan(math.pi / 4), 1.0)
+assert isclose(math.tan(-math.pi / 4), -1.0)
+
+# asin - arc sine (inverse sine)
+assert isclose(math.asin(0), 0.0)
+assert isclose(math.asin(1), math.pi / 2)
+assert isclose(math.asin(-1), -math.pi / 2)
+assert isclose(math.asin(0.5), math.pi / 6)
+
+# acos - arc cosine (inverse cosine)
+assert isclose(math.acos(1), 0.0)
+assert isclose(math.acos(0), math.pi / 2)
+assert isclose(math.acos(-1), math.pi)
+assert isclose(math.acos(0.5), math.pi / 3)
+
+# atan - arc tangent (inverse tangent)
+assert isclose(math.atan(0), 0.0)
+assert isclose(math.atan(1), math.pi / 4)
+assert isclose(math.atan(-1), -math.pi / 4)
+
+# atan2 - arc tangent of y/x (handles quadrants correctly)
+assert isclose(math.atan2(0, 1), 0.0)
+assert isclose(math.atan2(1, 0), math.pi / 2)
+assert isclose(math.atan2(0, -1), math.pi)
+assert isclose(math.atan2(-1, 0), -math.pi / 2)
+assert isclose(math.atan2(1, 1), math.pi / 4)
+assert isclose(math.atan2(-1, -1), -3 * math.pi / 4)
+
+# ============================================================
+# Angle conversion functions: degrees, radians
+# ============================================================
+
+# degrees - convert radians to degrees
+assert isclose(math.degrees(0), 0.0)
+assert isclose(math.degrees(math.pi), 180.0)
+assert isclose(math.degrees(math.pi / 2), 90.0)
+assert isclose(math.degrees(2 * math.pi), 360.0)
+
+# radians - convert degrees to radians
+assert isclose(math.radians(0), 0.0)
+assert isclose(math.radians(180), math.pi)
+assert isclose(math.radians(90), math.pi / 2)
+assert isclose(math.radians(360), 2 * math.pi)
+
+# ============================================================
+# Arithmetic functions: fmod, modf, copysign, fsum, gcd, factorial
+# ============================================================
+
+# fmod - floating point modulo
+assert isclose(math.fmod(5.0, 3.0), 2.0)
+assert isclose(math.fmod(-5.0, 3.0), -2.0)
+assert isclose(math.fmod(5.0, -3.0), 2.0)
+assert isclose(math.fmod(-5.0, -3.0), -2.0)
+assert isclose(math.fmod(4.0, 3.0), 1.0)
+assert isclose(math.fmod(-4.0, 3.0), -1.0)
+assert isclose(math.fmod(-2.0, 3.0), -2.0)
+assert isclose(math.fmod(2.0, 3.0), 2.0)
+
+# modf - split into fractional and integer parts
+frac, integer = math.modf(1.5)
+assert isclose(frac, 0.5)
+assert isclose(integer, 1.0)
+
+frac, integer = math.modf(-1.5)
+assert isclose(frac, -0.5)
+assert isclose(integer, -1.0)
+
+frac, integer = math.modf(2.0)
+assert isclose(frac, 0.0)
+assert isclose(integer, 2.0)
+
+frac, integer = math.modf(0.0)
+assert isclose(frac, 0.0)
+assert isclose(integer, 0.0)
+
+# copysign - return x with the sign of y
+assert isclose(math.copysign(1.0, -1.0), -1.0)
+assert isclose(math.copysign(-1.0, 1.0), 1.0)
+assert isclose(math.copysign(1.0, 1.0), 1.0)
+assert isclose(math.copysign(-1.0, -1.0), -1.0)
+assert isclose(math.copysign(5.0, -0.0), -5.0)
+assert isclose(math.copysign(0.0, -1.0), 0.0)  # sign of zero may vary
+
+# fsum - accurate floating point sum
 assert math.fsum([0.1] * 10) == 1.0
+assert math.fsum([]) == 0.0
+assert math.fsum([1.0, 2.0, 3.0]) == 6.0
+assert math.fsum([0.1, 0.2, 0.3]) == 0.6
 
-# test gcd
+# gcd - greatest common divisor
 assert math.gcd(10, 5) == 5
 assert math.gcd(10, 6) == 2
 assert math.gcd(10, 7) == 1
 assert math.gcd(10, 10) == 10
 assert math.gcd(-10, 10) == 10
+assert math.gcd(10, -10) == 10
+assert math.gcd(-10, -10) == 10
+assert math.gcd(0, 5) == 5
+assert math.gcd(5, 0) == 5
+assert math.gcd(0, 0) == 0
 
-# test fmod
-assert math.fmod(-2.0, 3.0) == -2.0
-assert math.fmod(2.0, 3.0) == 2.0
-assert math.fmod(4.0, 3.0) == 1.0
-assert math.fmod(-4.0, 3.0) == -1.0
-
-# test modf
-x, y = math.modf(1.5)
-assert isclose(x, 0.5)
-assert isclose(y, 1.0)
-
-x, y = math.modf(-1.5)
-assert isclose(x, -0.5)
-assert isclose(y, -1.0)
-
-# test factorial
+# factorial - n!
 assert math.factorial(0) == 1
 assert math.factorial(1) == 1
 assert math.factorial(2) == 2
 assert math.factorial(3) == 6
 assert math.factorial(4) == 24
 assert math.factorial(5) == 120
+assert math.factorial(10) == 3628800
 
+# ============================================================
+# Special value tests
+# ============================================================
+
+# Test NaN generation from invalid operations
+a = -0.1
+a = a ** a
+assert math.isnan(a)
+assert not math.isinf(a)
+
+# Test infinity
+assert math.isinf(float("inf"))
+assert math.isinf(float("-inf"))
+assert not math.isnan(float("inf"))
\ No newline at end of file
diff --git a/tests/930_deterministic_float.py b/tests/930_deterministic_float.py
index dc024b0c..13522584 100644
--- a/tests/930_deterministic_float.py
+++ b/tests/930_deterministic_float.py
@@ -41,15 +41,15 @@ assertEqual(math.gcd(10, 10), 10)
 assertEqual(math.gcd(-10, 10), 10)
 
 # test isfinite, isinf, isnan
-assertEqual(math.isfinite(math.pi), 1)
-assertEqual(math.isfinite(math.inf), 0)
-assertEqual(math.isfinite(math.nan), 0)
-assertEqual(math.isinf(math.pi), 0)
-assertEqual(math.isinf(math.inf), 1)
-assertEqual(math.isinf(math.nan), 0)
-assertEqual(math.isnan(math.pi), 0)
-assertEqual(math.isnan(math.inf), 0)
-assertEqual(math.isnan(math.nan), 1)
+assertEqual(math.isfinite(math.pi), True)
+assertEqual(math.isfinite(math.inf), False)
+assertEqual(math.isfinite(math.nan), False)
+assertEqual(math.isinf(math.pi), False)
+assertEqual(math.isinf(math.inf), True)
+assertEqual(math.isinf(math.nan), False)
+assertEqual(math.isnan(math.pi), False)
+assertEqual(math.isnan(math.inf), False)
+assertEqual(math.isnan(math.nan), True)
 
 # test exp
 assertEqual(math.exp(0), 1.0)