is there other ways to handle text than jit.gl.text?

Two things to say:

First, I was wrong when I said earlier that the helpfile for jit.gl.textmult was not working. In fact it was not working for me, as it appears to be a bug with only some GPUs (in my case the integrated graphics Intel UHD 630). I relaunched Max with the dedicated GPU, and tried on another computer with another GPU, and the helpfile works as expected with those, so sorry if that misleaded you as you most certainly did not encounter that issue! I submitted the issue and it's being investigated.

Then:

However, upon a quick look at the jit.gl.textmult script , it appears that it may serve a different purpose.

I am not so sure. Here are some heavily truncated extracts from multiple_text.js found inside of [jit.gl.textmult], with some extra comments I added:

function TextObj() 
{   
    this.glText = new JitterObject("jit.gl.text"); //THIS LINE CREATES A NEW JIT.GL.TEXT
    //[...]
}
//[...]
function SetTextAndDrawBang() //THIS FUNCTION IS CALLED WHENEVER A NEW TEXT IS RECEIVED
{   
    //[...]
    if (Array.isArray(g_text)) {
        //[...]
        for (var i=0; i<g_text.length; i++) { 
            //[...]
            CreateTextObjs(); //THIS FUNCTION (DETAILED BELOW) IS CALLED FOR EACH ELEMENT OF TEXT (ie each element of the array g_text)
            //[...]
            g_textObj.SetText(g_text[i]);
            g_textObj.Draw();
            g_wordIndex++;
        }
    //[...]
}
//[...]

function SetAllTextAttributes()
{   
    if (g_textObj != null) {
        g_textObj.SetMode(mode); // THIS MIGHT BE WHERE YOU CAN MANUALLY SET THE MODE TO "2d"
        // g_textObj.SetFont(font);
        g_textObj.SetDepth(depth)
        g_textObj.SetWeight(weight);
        g_textObj.SetAlign(align);
    }

} 

function CreateTextObjs()
{    
    if (g_textObj == null) {
        g_textObj = new TextObj(); // IT CALLS TextObj() (see at the top) WHICH CREATES A NEW JIT.GL.TEXT
        SetAllTextAttributes();
    }
}

So yes, that js file indeed creates a new [jit.gl.text] for each element of text.

There is also a lot of code that deals with the control matrices (for position, scale, color...), but I'm still struggling to get what's going on, and what makes the 2d mode to not work. But that problem hooked I guess so I might investigate a bit further.

Your specific case would require a lot of work to modify the existing [jit.gl.textmult] that would allow you to first instanciate a given number of [jit.gl.text], even with empty text, then dynamically populate them individually with your text as it comes. It should make the generation much faster.