Predict glucose. With the evidence behind it.
January AI and SNAQ both predict your glucose response from a photo. SNAQ adds a peer-reviewed RCT on Time-in-Range and dedicated GLP-1 guidance that January AI doesn't ship today.
In short.
January AI predicts glucose response from a photo without a CGM, and was recognised on TIME's Best Inventions 2025 list. It's positioned for prediabetes and T2D not on insulin. SNAQ does the same predictive job and adds two things January cannot offer today: a peer-reviewed RCT in eClinicalMedicine showing a +6.6 pp Time-in-Range improvement, and dedicated GLP-1 guidance. The functional overlap is real; the evidence and cohort coverage are not.
At a glance.
- January AI is the closest functional twin to SNAQ on predictive meal guidance, with TIME Best Inventions 2025 recognition.
- SNAQ has a peer-reviewed RCT (eClinicalMedicine 2025, +6.6 pp Time-in-Range). January AI has a retrospective cohort study in npj Digital Medicine.
- SNAQ's RCT was conducted in T1D adults on automated insulin delivery; January AI is positioned for prediabetes and T2D not on insulin.
- SNAQ integrates with 20+ devices. January AI's integration set is narrower.
- SNAQ offers a 7-day free trial plus a 30-day money-back guarantee on web purchases; January AI is subscription-only.
Feature by feature.
Structured comparison based on publicly available information, last reviewed May 2026.
| Feature | SNAQ | January AI |
|---|---|---|
| Glucose prediction without a CGM | ✅ | ✅ |
| Photo meal recognition | ✅ Validated at 5.5 g mean absolute carb error | ✅ Large food database, accuracy not peer-reviewed |
| Peer-reviewed clinical evidence (product-specific) | ✅ RCT, eClinicalMedicine 2025 (+6.6 pp TIR) | ⚠ Retrospective cohort, npj Digital Medicine |
| GLP-1 protein intake and muscle-preservation | ✅ Dedicated | ❌ |
| Device coverage | ✅ 20+ devices (Dexcom, FreeStyle Libre, Stelo, Lingo and more) | ⚠ Narrower set |
| Provider dashboard for clinicians | ✅ SNAQ Care, 500+ clinicians | ❌ |
| Best for | Anyone whose glucose matters: T1D, T2D, prediabetes, GLP-1 | Prediabetes and T2D wellness, sensor-bundle preference |
Sources: Herzig D. et al., JMIR mHealth and uHealth, 2020 (carb counting accuracy) Piazza et al., eClinicalMedicine 2025 (RCT NCT05671679, Time-in-Range) SNAQ research overview January AI on TIME Best Inventions 2025 January AI product pages
Why users switch.
RCT over retrospective cohort
January AI's evidence is a retrospective cohort study. SNAQ's is a randomised controlled trial in T1D users on automated insulin delivery.
Bring whichever sensor you already wear
Dexcom, FreeStyle Libre, Stelo, Lingo, Eversense and more. No sensor lock-in, no bundled subscription.
Three cohorts in one app
T1D, T2D and GLP-1 users get distinct guidance, including muscle-preservation protein flagging for GLP-1.
Honest take.
Choose SNAQ if…
- You want peer-reviewed RCT evidence behind your predictive tool.
- You take a GLP-1 medication, or want guidance built around T1D, T2D or prediabetes nutrition.
- You want a low-friction way to try, and the option to bring your own sensor.
Consider January AI if…
- You're new to CGM and want a guided, sensor-bundled onboarding.
- You prefer a single subscription that bundles sensor and app together.
Bottom line
Both apps predict glucose. SNAQ is the one with peer-reviewed RCT evidence and dedicated GLP-1 guidance.
Published randomised controlled trial evidence, measured against standard care in T1D patients on automated insulin delivery systems.
Source: eClinicalMedicine (The Lancet), 2025 · Piazza et al. · NCT05671679
Frequently asked.
What's the main difference between SNAQ and January AI?
Is the clinical evidence for SNAQ independently verified?
Which CGMs and devices does SNAQ work with?
How is SNAQ different from January AI for T1D users?
How is SNAQ's clinical evidence different from January AI's?
See the difference yourself.
No logbook can coach you. Download SNAQ free and find out what your data has been trying to tell you.
4.6 · 3,000+ App Store ratings