Can we measure $Delta G$ when a reaction hasn't reach equilibrium?

I am asking because we say that thermodynamic variables can only calculated at equilibrium. So the above states ($1$ and $2$) must be equilibrium states. But they aren't as the reaction still keeps going.

They are in chemical equilibrium as long as the forward and reverse reaction rates are equal.

It's the same for thermal equilibrium. Two objects are in thermal equilibrium with one another if they are at the same temperature. But even though they are in thermal equilibrium, at the microscopic level energy continues to transfer between particles of the two objects. It is just that the rate of energy transfer from particles in object 1 to particles in object 2 equals the rate of energy transfer from particles of object 2 to particles of object 1 for a net transfer of energy of zero resulting in thermal equilibrium.

Hope this helps.

My answer is based on Astarita: Thermodynamics: An Advanced Textbook for Chemical Engineers (Springer 1990), especially chapters 2 and 3 (Homogeneous reactions). There, the answer to your question is: it depends.

The free energy $G$ (and also entropy and other kinds of free energy) is defined also out of equilibrium.

There are systems for which the free energy depends, out of equilibrium, only on the same variables that define the equilibrium of the system (Astarita calls these the "site"). For example let's say temperature $T$ and volume $V$. Then $G(t) = G[T(t), V(t)]$ at every time $t$. Let's say that at some time $t_0$ the system passes through values $T(t_0)=T_0$ and $V(t_0) = V_0$, without being in equilibrium. Then you can measure the free energy at time $t_0$, even if the system is not at equilibrium then, and the value you find is the same the system would have in equilibrium at $T_0$ and $V_0$.

There are systems for which the free energy depends also on non-equilibrium variables. For example we could have $G(t) = G[T(t), V(t), dot{V}(t)]$, where $dot{V}(t)$ is the instantaneous rate of change of volume at time $t$. In this case, even if $T(t_0)=T_0$ and $V(t_0) = V_0$ at some time $t_0$, the free energy doesn't have, at that time, the value it would have in equilibrium at $T_0$ and $V_0$ (which is $G[T=T_0, V=V_0, dot{V}=0]$), because it depends on $dot{V}(t_0)$, which is different from zero if the system is out of equilibrium.

What I wrote is just an example; the dependence could be on other non-equilibrium quantities and rates of change.

It's worth taking a look at the book, because it discusses such non-equilibrium matters, especially in regard to chemical reactions, at great lengths.