M = 10^(-3); \[Epsilon] = 10^(-5); gy = 0.35;
g = N[Sqrt[(4 Pi)/30], 3]; e = N[Sqrt[(4 Pi)/129], 3]; f := - gy; sw =
Sqrt[0.23]; cw = Sqrt[1 - 0.23];
Mb[j_] := Mblist[[j]];
gb[j_] := gblist[[j]];
Mblist = {0, 93};
gblist = {e, g};
(*f={e,\[Mu],\[Tau],\[Nu],\[Nu],\[Nu],u,c,t,d,s,b},1~3E,4~6\[Nu],7~9U,\
10~12D*)
mf[i_] := mflist[[i]];
Qf[i_] := Qflist[[i]];
Subscript[Qfb, 1][i_] := Qfblist1[[i]];
Subscript[Qfb, 2][i_] := Qfblist2[[i]];
mflist = {511 10^-6, 106 10^-3, 1.8, 0, 0, 0, 2.3 10^-3, 1.3, 173,
4.8 10^-3, 95 10^-3, 4.2};
Qflist = {1, 1, 1, 2/3, 2/3, 2/3, 1/3, 1/3, 1/3};
Qfblist1 = {-1, -1, -1, 2/3, 2/3, 2/3, -(1/3), -(1/3), -(1/3)};
t[s_, p_, i_, j_] :=
1/2 (M^2 + Mb[j]^2 - s) +
2 p Sqrt[(s/4 - mf[i]^2) ((1/2 (Sqrt[s] + M^2/Sqrt[s]))^2 - M^2)];
u[s_, p_, i_, j_] :=
1/2 (M^2 + Mb[j]^2 - s) -
2 p Sqrt[(s/4 - mf[i]^2) ((1/2 (Sqrt[s] + M^2/Sqrt[s]))^2 - M^2)];
F1[s_, i_, j_] := (Qf[i] f Subscript[Qfb, j][i] gb[j])^2/(32 Pi s)
Sqrt[((s - (M + Mb[j])^2) (s - (M - Mb[j])^2))/(
s (s - 4 mf[i]^2))] ;
A[s_, p_, i_,
j_] := (-2 (2 mf[i]^2 + Mb[j]^2) (2 mf[i]^2 + M^2) (1/
t[s, p, i, j]^2 + 1/u[s, p, i, j]^2) +
2 (t[s, p, i, j]/u[s, p, i, j] + u[s, p, i, j]/t[s, p, i, j]) -
4 (1/u[s, p, i, j] + 1/t[s, p, i, j]) (2 mf[i]^2 + M^2 +
Mb[j]^2) -
4/(u[s, p, i, j] t[s, p, i, j]) (2 mf[i]^2 + M^2 +
Mb[j]^2) (2 mf[i]^2 - M^2 - Mb[j]^2)) ;
d[i_, j_] := Max[M + Mb[j], 2 *mf[i]];
G[t_] := NIntegrate[
F1[s, 1, 1] A[s, p, 1, 1] (s - d[1, 1]^2) s^(1/2)
t^3 BesselK[1, s^(1/2) t/M], {p, -1, 1}, {s, 4 mf[1]^2,
Infinity}, WorkingPrecision -> 16, MinRecursion -> 5,
MaxRecursion -> 1000, AccuracyGoal -> Infinity,
Method -> {Automatic, "SymbolicProcessing" -> 0}];
F[x_] := NIntegrate[G[t], {t, 0, x}, WorkingPrecision -> 16,
MinRecursion -> 5, MaxRecursion -> 1000, AccuracyGoal -> Infinity,
Method -> {Automatic, "SymbolicProcessing" -> 0}];
结果如图

s没有值

F[x_] := NIntegrate[
F1[s, 1, 1] A[s, p, 1, 1] (s - d[1, 1]^2) s^(1/2) t^3 BesselK[1,
s^(1/2) t/M], {t, 0, x}, {p, -1, 1}, {s, 4 mf[1]^2, Infinity},
WorkingPrecision -> 16, MinRecursion -> 5, MaxRecursion -> 1000,
AccuracyGoal -> Infinity,
Method -> {Automatic, "SymbolicProcessing" -> 0}];
其余不变
最后输出结果为
@ asdasd1dsadsa

这个问题比较复杂，我已经定位到了，待会跟你解释。

你是不是会错意了。
我的意思是，你的错误相当于：
u[x_] := 1
v[x_] := u[x]
Integrate[v[x]*u, {u, 0, 1}, {x, 0, 1}]