CamAPS FX

System profile

What it looks like

Both are tubed pumps; the reservoir sits inside the pump body and tubing runs to a cannula site. The algorithm itself runs in the CamAPS FX app on the phone, not on the pump.

Two pumps, one app: YpsoPump and Dana-i

CamAPS FX is unusual among current AID systems in offering a genuine choice of pump hardware. The same Cambridge algorithm runs on either the YpsoPump or the Dana-i, both tubed pumps, and in both cases the algorithm lives on the wearer’s Android phone or iPhone in the CamAPS FX app, not on the pump. The YpsoPump-plus-CamAPS bundle is marketed by Ypsomed as mylife CamAPS, which is why you will see the same system described under both names. Whichever pump you are on, everything in this guide about how CamAPS FX behaves applies equally; the difference is in the hardware, not the algorithm.

CGM interoperability

CamAPS FX pairs with the Dexcom G6, the established sensor for both the YpsoPump and the Dana-i configurations, and with the newer Dexcom G7 (rolling out with CamAPS FX from summer 2026), plus the FreeStyle Libre 3 Plus (the AID-paired Libre variant; the standard Libre 3 does not currently pair). Regional availability varies, so verify the current market pairing before a market-specific claim.

Age bands and regulatory status

How the algorithm works

CamAPS FX runs the Cambridge model predictive control (MPC) algorithm, and it is the one system covered here that adapts to you over days and weeks. Where Control-IQ applies a fixed set of rules and 780G tunes around a target you set, CamAPS FX builds a rolling forecast of where your glucose is heading, compares it against your chosen target, and adjusts insulin delivery to steer towards that forecast. Crucially, it also adapts to you in the background over days and weeks, adjusting around your individual day-to-day insulin needs and circadian variation, so the same settings behave differently, and usually better, a month in than on day one (Tauschmann 2018, Lancet; Boughton 2026, Metabologia, n=35,714, median TIR 69.6%). The algorithm itself lives on the CamAPS FX app on the wearer’s Android phone or iPhone, not on the pump: the phone does the forecasting and the pump delivers.

The personal glucose target is the primary lever. The target is the number you and your team set to steer control tighter or looser, and it is what CamAPS FX gives you to adjust across the day, in separate periods, from 4.4 to 11.0 mmol/L (80 to 200 mg/dL) in 0.1 mmol/L steps, default 5.8 mmol/L. This is the setting you reach for session to session: it is the lever you turn, and lower personal targets are associated with more time in range (Ware 2025).

The insulin-to-carbohydrate ratio is a secondary lever, not a direct input to the algorithm. A weaker ICR leaves more of the work to algorithm-driven insulin; a stronger one pulls delivery back towards the bolus insulin you command. The important difference from the other systems is that CamAPS FX does not give you an adjustable insulin-action-time setting the way 780G does; the loop runs its own insulin model, which you cannot shorten or lengthen, and that model is in fact the closest to real physiology of the current systems. The practical consequence is that an ICR set too strong, too much insulin per gram of carbohydrate, is not something you can offset elsewhere: the extra insulin you bolus sits in the system as active insulin the algorithm has to account for and work around, and it cannot simply discount insulin you have already delivered. So be cautious with tight insulin-to-carb ratios; the setting to focus on is the target, not the ICR.

Boost and Ease-off are event-level overrides that sit on top of the chosen target, and both have real-world safety data behind them (Royston and Hovorka 2024, DTT, n=7,464). Boost tells the algorithm to assume roughly 35% higher insulin needs, for illness, growth surges, persistent highs, or pregnancy; during Boost, time below 3.9 mmol/L was 0.0%, with no severe hypoglycaemia or DKA attributed to it, because the algorithm stays glucose-responsive and eases off as glucose falls. Ease-off reduces delivery, for exercise, post-exercise hypo risk, or increased insulin sensitivity; counterintuitively, time above range fell during Ease-off (15% versus 25% outside it), reflecting the exercise-glucose interaction rather than any loss of control. The clinical message: Boost does not cause hypoglycaemia, and Ease-off does not cause hyperglycaemia. Both are most used by caregivers of young children.

Exercise: getting the most out of Ease-off mode

Ease-off is the exercise mode. It reduces insulin delivery for exercise, post-exercise hypo risk, and periods of higher insulin sensitivity. In real-world data, time above range actually fell during Ease-off (15% versus 25% outside it), reflecting the exercise-glucose interaction rather than any loss of control, and there was no increase in time low (Royston and Hovorka 2024, n=7,464). As with any AID system, activating it ahead of the session, rather than as you start, gives the algorithm time to reduce the insulin already on board.

Then take small carbohydrate top-ups every 20 minutes as needed, rather than one large amount upfront: a big carbohydrate load can spike glucose and prompt a stronger response, raising hypo risk later in the session.

This is a live capture of GNL’s free Carbs for Thirty Minutes of Exercise Explorer, one of the tools on the Grace page. To find it: open theglucoseneverlies.com/gnl-grace, accept the one-off consent step, choose the Explorers tab, then Carbs for Thirty Minutes of Exercise. The snippet below shows the input side (top: 50 kg, 5.5 mmol/L falling slowly, aerobic exercise, AID/auto pump insulin therapy, a 7-unit bolus taken 90 minutes earlier) and the output it produced (bottom: insulin-on-board and the estimated carbohydrate range) for a 50 kg person. Enter your own weight and recent doses to get your own estimate.

For planned, longer sessions, the T25/T25 framework below is written for AID systems specifically; the manual reductions are smaller than the injections or pump-only version because the algorithm is already doing some of the work.

The AID and Exercise Consensus Guideline (EASD/ISPAD 2025, Moser, Zaharieva, Pemberton et al.) is worth reading alongside this.

Hypo management

The algorithm prevents many overnight hypos by easing or suspending insulin as glucose falls or is forecast to fall, but lows still happen, especially with unplanned exercise or a bolus that turns out too strong for the situation. Treatment amounts often need to be smaller than before AID, because CamAPS FX has usually already reduced delivery earlier than you would expect. Recording the treatment in the Add Meal menu as a hypoglycaemia treatment, rather than as a normal meal, tells the algorithm those carbohydrates are corrective, so the loop does not stack more insulin on top of them.

This is a live capture of GNL’s free Hypo Treatment Explorer, another of the tools on the Grace page. To find it: open theglucoseneverlies.com/gnl-grace, accept the one-off consent step, choose the Explorers tab, then Hypo Treatment Explorer. The snippet below shows the input side (top: 50 kg, 3.4 mmol/L, falling slowly) and the output it produced (bottom: a fast-acting carbohydrate estimate with everyday food options). Enter your own weight, glucose, and trend for your own estimate.

Under-5s

CamAPS FX carries the widest paediatric indication of any AID system, licensed from age 1, and the low adjustable target with Boost and Ease-off is what makes it workable for the unpredictable eating and pronounced overnight glucose swings of infants and pre-schoolers. Because the target floor goes as low as 4.4 mmol/L and moves in 0.1 mmol/L steps across separate periods of the day, families can steer tighter overnight and looser through the day with their diabetes team.

Children under six show a distinct diurnal insulin-need pattern: a clear evening rise in insulin need followed by an overnight fall (Biester et al. 2023, DPV registry, n=25,718). On CamAPS FX, the lever families use with their team is the personal target across those windows, tighter overnight and easing through the evening rise, the same dusk-then-dawn pattern that shapes preschool settings on every AID system, not just this one. This is a clinic conversation, not a parental adjustment. GNL covers that overnight “dusk till dawn” pattern in the under-5s in detail, with real before-and-after CGM traces, in the infants and pre-schoolers guide.

Notes