The Best Melatonin Alternatives: What the Research Says About Each

Why people look for an alternative

Melatonin is a timing signal. It is best supported for jet lag, and NIH says there is not enough strong evidence to recommend it for long-term trouble sleeping. Many people also want to avoid a nightly hormone, dislike the dose uncertainty (lab tests found gummies at 74% to 347% of their label, per Cohen et al., 2023), or find that what keeps them awake is stress rather than timing. Melatonin side effects covers the research in detail.

The options at a glance

The first choice: CBT-I

Cognitive behavioral therapy for sleep is a short, structured program: the bed-is-for-sleep rule, a fixed wake time, a period of limited time in bed, and work on the thoughts that keep you awake. In a meta-analysis of 20 trials it cut the time to fall asleep by about 19 minutes and time awake at night by about 26 minutes compared with no treatment (Trauer et al., 2015), and the American Academy of Sleep Medicine strongly recommends it as the first approach for long-term trouble sleeping (Edinger et al., 2021). It is available through apps, online courses and specialists. Several of its techniques are in how to fall asleep fast.

L-theanine

An amino acid found in tea. In a four-week crossover trial of 30 healthy adults, 200 mg a day improved sleep-quality scores compared with placebo (Hidese et al., 2019). A 2026 meta-analysis of 31 trials found a modest effect on stress, partly driven by lower-quality studies, and a clearer effect on reaction time (Gerolymos et al., 2026). A 2026 review reports that 200 to 450 mg a day supports sleep quality without sedation (Cotter et al., 2026). More in L-theanine vs α-casozepine.

Oral lavender oil

One standardized oral lavender oil preparation beat placebo in five 10-week trials of 1,213 people on scores for tension and worry (Dold et al., 2023). In a re-analysis of one trial, almost all of its effect on sleep, 98%, ran through that calming effect rather than sedation (Seifritz et al., 2019). The limits: the trials were in people with a diagnosed condition, not everyday stress, they tested one specific product, and most were funded by its maker.

Ashwagandha

An herb used in Ayurvedic tradition. A meta-analysis of 5 placebo-controlled trials with 400 participants found a small but significant benefit for overall sleep, clearest at 600 mg a day or more for at least 8 weeks (Cheah et al., 2021). In a safety trial of 1,002 adults, 600 mg a day for 8 weeks caused no changes in liver tests compared with placebo (Pakhale et al., 2026).

But it carries a rare, serious risk. A 2026 review of published cases found 25 people with liver injury linked to ashwagandha, usually after weeks of use. Most recovered after stopping, but the outcomes included one liver transplant and three deaths in people who already had serious liver disease (McIntyre et al., 2026). If you take it, stop and see a doctor if you notice yellowing skin or eyes, itching or dark urine, and avoid it if you have any liver problem.

The limited-evidence group

• Chamomile: a small pooled analysis found better sleep-quality scores (Hieu et al., 2019), but a placebo-controlled trial in adults with long-term trouble sleeping found no difference in sleep diaries (Zick et al., 2011).
• Passionflower: in a 2-week trial of 110 adults, extract added about 23 minutes of laboratory-measured sleep compared with placebo, a borderline result, and other measures did not beat placebo (Lee et al., 2020).
• Lemon balm: two small placebo-controlled trials were positive (Di Pierro et al., 2024; Bano et al., 2023) but have not been repeated.
• Glycine: 3 g at bedtime reduced next-day fatigue after short nights in a very small study (Bannai et al., 2012).

Popular, but the evidence falls short

• Magnesium: a 2026 review of 12 trials found inconsistent, low-certainty evidence and did not support routine use for sleep (Lopresti et al., 2026). One small trial in older adults was positive (Abbasi et al., 2012). See magnesium vs α-casozepine.
• Valerian: reviews of the research found no objective evidence that it improves sleep (Valente et al., 2024; Ell et al., 2023), and it can add to the effect of sedating medicines (Mielczarek et al., 2026).
• Tart cherry: a small pilot helped time awake at night (Pigeon et al., 2010); a later trial found no effect on any sleep measure (Tucker et al., 2024).
• GABA supplements: a 95-day placebo-controlled trial found no effect on sleep quality, measured as a secondary outcome (de Bie et al., 2023).

Where α-casozepine fits, and why we chose it

α-casozepine is a peptide from milk protein that works with GABA-A receptors, the brain's main calming system. Its evidence base is small: five human trials. At 150 mg a day for 30 days it reduced everyday stress symptoms compared with placebo (Kim et al., 2007), and at 300 to 600 mg a day it shortened laboratory-measured time to fall asleep compared with placebo (Chang et al., 2024). In another trial some measures improved while others did not beat placebo (Kim et al., 2019). No trial has reported a pattern of side effects.

We chose it because it acts on the calming signal directly rather than the body clock, it is not a hormone or a sedative, it has been studied for both stress and sleep, and its safety record is clean. We would rather build on a small, clean evidence base and say so than make claims the research does not support. What is α-casozepine? lists every trial and its limits.

Whichever you try, check with your doctor first if you take prescription medicine, are pregnant or breastfeeding, or have a long-term health condition.