Activation curves for different types of Insulin

This handy chart shows compared activation curves for Novolog, Humalog, (Novolin) R, NPH (Humulin/ Novolin N), Lantus, Levemir, human and analog mixtures, Tresiba, R500, Toujeo and Afrezza.

courtesy GroupHealth, heavily edited and added to by @Kmichel

A partial original document is here.

Hopefully, this can serve as a reference in the future. As always, these charts are generic and indicative only: they may not reflect what happens to an individual user.

[EDIT] This post has been modified multiple times as the chart has evolved.

End of wiki ---------- comments start here


I like the chart. Very useful. I am surprised in a web search I could not find a chart that included some of the newer insulins in it. Here is a link to a table that claims it is a chart (but isn’t). I guess I will have to do an actual chart for it.


I’m no photoshop expert, obviously.


I thought Toujeo worked the same as Lantus, just in a more potent concentration? Is that wrong?

Yes - my understanding that Toujeo has a longer duration curve than lantus - here is from the product sheet. (This assumes that you reached a steady state of Toujeo from previous days dosage so it is not the same as in the chart in the top post.)


Kevin, that was a great idea!!!

If you want send me the pics, I can try to do as seamless as job as possible, and post it at the top of the thread, with acknowledgment to you of course.!

1 Like

I wonder if that’s actually supposed to be different from Lantus or just looks different because of the assumptions that it’s “multiple doses” in the toujeo profile and not in the Lantus

In my understanding toujeo is just u300 Lantus, I cant conceptuslize how it would last any longer

1 Like

I was also surprised last year when I read about it, after researching what @docslotnick was writing about, but the sources seemed impeccable then, so, although it seems counterintuitive, I believe it is correct.

1 Like

I have no pictures other than what I posted. It was a cut, stretch, shrink, and paste job done on a screen capture :wink:

I also take these curves with a grain of salt because if you look at the afrezza curve it appears to be just marginally faster than humalog and I know with absolute certainty that this nowhere near reflects the reality


Dunno. I just went off the data in the chart I posted above it. I didn’t verify that data.

Also, the chart claims 15 minutes onset for Afrezza, and that isn’t my experience with it.

If I did the whole chart from scratch, rather than starting from the original one, I wouldn’t have used a 2 hour time scale. Maybe I’ll have at ring it while I’m on the plane tonight.


Oh! I didn’t even actually see afrezza on your table on my cell phone… I just meant the normal curves I’ve seen floating around in the past make it look way more similar to injected bolus that it actually is… your picture is better than they are… though still I think it’s a bit goofy when they say injected “starts working in 15 minutes” yeah, right, maybe in some infinitesimally small barely measurable way. Afrezza is up to full speed in 15 minutes…


It was a great job! I posted it to the original post. If you revisit it, I’ll make sure to repost it as well.

You had an inspiration there: this is the best chart of its kind that I have seen. It is tremendously useful, now that it is reasonably complete. We may need to add FIASP to it :slight_smile:


Yep. If it wasn’t such a pain to create the chart in their format, I would have made one with a time scale smaller than 2 hours granularity.

Maybe while I’m on the plane…


That would be nice.

@kmichel, it was a really good idea.