Ginkgo Biloba: What the Evidence Actually Shows

By Peter Benson, Cognitive Enhancement Researcher  ·  Updated October 2026  ·  Citations re-verified

Short answer: Ginkgo biloba is one of the best-tested “brain” supplements, and the results are mostly negative. Two large, long trials found it didn’t prevent dementia, and a meta-analysis found no memory benefit in healthy people. There may be a modest effect in existing dementia, but the evidence is inconsistent.

Ginkgo biloba is one of the best-selling memory supplements, and also one of the most thoroughly tested. That makes it a rare case where the question “does it work?” has a fairly clear answer, and the answer is narrower than the marketing.

This guide covers why ginkgo was expected to help, the large prevention trials, the evidence in healthy people and in dementia, safety, and what to look for if you try it. It sits in the Nootropics & Supplements hub.

Why was ginkgo expected to work?

Standardised ginkgo leaf extract contains flavonoids and compounds called terpene lactones. In laboratory and small human studies it has shown antioxidant effects, effects on blood flow and blood-vessel function, and blocking of platelet-activating factor, which is also the basis of the bleeding concern. Better circulation plus antioxidant protection was a plausible story for an ageing brain.

But plausible mechanisms often fail to change outcomes people notice. The way to know is large, long trials that measure memory and dementia directly, and ginkgo has had several.

Does ginkgo biloba prevent dementia?

No, on the best evidence. The Ginkgo Evaluation of Memory (GEM) study randomised 3,069 US adults aged 75 or older, with normal thinking or mild cognitive impairment, to 120mg of ginkgo extract twice a day or placebo, and followed them for about six years. 277 people on ginkgo and 246 on placebo developed dementia; ginkgo had no effect on overall dementia or Alzheimer’s (DeKosky et al., 2008; NIH). A further analysis found it didn’t slow cognitive decline either (Snitz et al., 2009). The study was NIH-funded, with the extract supplied by its manufacturer.

In France, the GuidAge trial gave standardised extract or placebo for five years to more than 2,800 older adults who had reported memory problems. 61 people on ginkgo and 73 on placebo developed Alzheimer’s, a difference that wasn’t statistically significant (Vellas et al., 2012). Two large, long trials, two negative results. For approaches with better evidence, see preventing cognitive decline and the Brain Health & Longevity hub.

Does ginkgo improve memory in healthy people?

This is how most ginkgo is sold, and it has the least support. A meta-analysis of 13 randomised trials in more than 1,000 healthy people found no effect on memory, attention or problem solving, whatever the age, dose or duration (Laws et al., 2012). If you’re healthy and hoping to feel sharper, expectations should be low. For realistic expectations, see what to expect from nootropics; Bacopa monnieri has more trial evidence for memory in healthy adults, though its effects are also modest.

Does ginkgo help people who already have dementia?

Possibly a little, but the evidence is mixed. Some randomised trials of the standardised extract EGb 761 at 240mg a day in people with mild-to-moderate dementia, particularly with behavioural symptoms, have reported modest improvements in selected cognitive, functional or behavioural outcomes (Herrschaft et al., 2012). Findings vary across studies, dementia types and outcomes, and much of the positive research is linked to the extract’s manufacturer. None of it shows that ginkgo slows the underlying disease or can replace approved treatments.

A Cochrane review of 36 trials concluded that ginkgo appears safe, with no more side effects than placebo, but that evidence of a predictable, clinically significant benefit for dementia or cognitive impairment was inconsistent and unreliable; of the four most recent trials, three found no difference from placebo (Birks & Grimley Evans, 2009). NCCIH likewise describes the evidence on symptoms as inconsistent (NCCIH). At most, this is something to discuss with a doctor as an add-on, never a replacement for established care.

Is ginkgo biloba safe?

General tolerability. NCCIH says leaf extracts are likely safe for most adults in moderate amounts and generally well tolerated; the most common side effects are dizziness, stomach symptoms and headache (NCCIH).

Bleeding. The evidence points both ways. A meta-analysis of 18 randomised trials found no increase in bleeding risk with standardised extracts (Kellermann & Kloft, 2011), but a review of 15 published case reports, eight involving bleeding in the brain, suggested a possible causal link, though most cases had other risk factors (Bent et al., 2005). Those trial results don’t establish that ginkgo is safe alongside anticoagulants or antiplatelet drugs. NCCIH’s guidance for health professionals states that taking ginkgo with warfarin is associated with an increased risk of major bleeding compared with warfarin alone, and advises against combining ginkgo with the HIV drug efavirenz (NCCIH, herb-drug interactions). If you take any blood-thinning medicine, consult your prescriber before using ginkgo, and tell your surgical team about all supplements before any operation.

Rodent cancer findings. In two-year studies by the US National Toxicology Program, very high doses of a ginkgo extract caused liver tumours in mice (“clear evidence” of carcinogenic activity) and thyroid tumours in rats (“some evidence”) (NTP, 2013). How this applies to people at normal doses is unknown, and industry groups disputed whether the extract matched clinical products, but it is a reason not to take ginkgo indefinitely without a clear benefit.

Seeds, seizures and pregnancy. Fresh ginkgo seeds are toxic, and serious effects have followed eating roasted seeds (NCCIH). There are rare reports of seizures in people with epilepsy taking ginkgo, so check with your doctor if you have a seizure disorder. Ginkgo may be unsafe in pregnancy and might cause early labour or extra bleeding if used near delivery (NCCIH). It may also interact with other medicines; ask a pharmacist.

Ginkgo biloba claims ranked by evidence

🟢 Stronger or replicated evidence  ·  🟡 Limited, mixed or preliminary evidence  ·  🔴 Not established or not supported

What dose did the trials use?

The large trials used a standardised leaf extract at 120mg twice a day, 240mg in total. Clinical extracts are standardised to roughly 24% flavone glycosides and 6% terpene lactones; unstandardised “ginkgo leaf powder” isn’t comparable. Trials that reported any effect in dementia ran for months, so there is no immediate effect to expect. These are research doses, not a recommendation.

What to look for if you try ginkgo

Look for a leaf extract standardised to about 24% flavone glycosides and 6% terpene lactones, with the amount per serving stated and independent testing. Avoid unstandardised leaf powder and never eat raw or roasted seeds. There are no shopping links here: ginkgo isn’t something I take, and I only link products I use.

Frequently asked questions

Scientific references