Metformin · dosing
Metformin dosage for longevity: every number is borrowed
No metformin dosage for longevity has been validated: the TAME design proposes 1,500 mg daily for 6 years, a figure borrowed from diabetes medicine. Nobody has tested a dose against an aging endpoint, and the animal lifespan evidence has been going the wrong way.
Where do the metformin dosage for longevity numbers come from?
Early / mixed evidence From diabetes practice, and the most concrete figure is the trial design: participants would take 1,500 mg daily over six years with multi-year follow-up.[1] That amount is a standard therapeutic dose from diabetes practice, chosen for familiarity and tolerability rather than derived by testing doses against an aging outcome.[2]
Clinics prescribing off-label mostly use the same range, typically starting at 500 mg and titrating toward 1,000 to 2,000 mg daily. Those are prescribing conventions carried over from a different indication, which is why any metformin dosage for longevity quoted online traces back to glucose control.
Show the numbers as a table
| Measure | Value |
|---|---|
| Typical starting dose | 500 mg/day |
| TAME design | 1,500 mg/day |
| Common off-label top | 2,000 mg/day |
| Above usual maximum | 3,000 mg |
Tracing each quoted figure to its origin
Laid side by side, the figures share one feature: each was set to control blood sugar or tolerability, and none was compared against another dose on an aging outcome.
| Quoted figure | Where it comes from | Tested against aging? |
|---|---|---|
| 500 mg daily | Usual diabetes starting dose, used to limit stomach upset | No |
| 1,500 mg daily | TAME design, a standard therapeutic amount | Planned, no results |
| 1,700 mg daily | MILES, a short study of gene expression | Markers only |
| 2,000 mg daily | Common upper figure in off-label prescribing | No |
Has any metformin dose been validated for aging?
No, because the effect it is supposed to produce has not been demonstrated. A validated dose requires an endpoint to titrate against, and for aging in healthy people there is none.
The evidence has also moved unfavorably: a 2025 review noted that metformin given to mice from mid-life failed to produce significant lifespan benefits.[3] A meta-analysis pooling 167 studies across vertebrate species found that dietary restriction and rapamycin extended lifespan consistently while metformin did not.[4]
How metformin is thought to act
The proposed mechanisms are many. Reviews describe effects on the mitochondrial respiratory chain, activation of AMPK, dampening of mTORC1 signaling, lysosomal pathways and the gut microbiome.[1] None of these has a known dose that separates an aging effect from a glucose effect.
The animal numbers that started the enthusiasm were also modest. A 2025 review put the mean lifespan increase reported across model organisms at 5.8%, before the less favorable mid-life mouse data.[3] A small average across species gives no basis for choosing between 500 mg and 2,000 mg in a person.
The human signal and its confounding
The human case rests largely on observational analyses suggesting people with diabetes taking metformin had mortality comparable to non-diabetic controls, a striking finding with obvious confounding.[5] No metformin dosage for longevity can be read out of that kind of data, since the doses were set for blood sugar.
Show the numbers as a table
| When | Event | Detail |
|---|---|---|
| 2016 | TAME rationale | Observational signal |
| 2019 | Exercise trial | Blunted adaptation |
| 2019 | TAME design review | 1,500 mg for 6 years |
| 2025 | Mouse review | No mid-life benefit |
| 2025 | Meta-analysis | No consistent effect |
What did human trials in older adults find?
Not much, and one result deserves more attention than it gets. A trial giving 500 mg three times daily for four months to older adults with probable sarcopenia and frailty did not improve physical performance.
That is a negative result at a full therapeutic dose in exactly the population longevity prescribing targets. It does not settle the question and it belongs in any honest discussion of dosing.
What the MILES study measured
The Metformin in Longevity Study (MILES) gave older adults 1,700 mg daily for several weeks and looked at gene expression in muscle and fat tissue.[1] It reported shifts in metabolic gene activity, which is a marker result rather than a health outcome.
It is often cited as proof that metformin reverses aging in humans. Read accurately, it shows the drug changes tissue signaling at a diabetes dose, which nobody disputed, and says nothing about whether a lower or higher dose would change how long anyone lives.
Does metformin blunt the benefits of exercise?
In older adults, yes: a randomized trial found that metformin attenuated the improvement in insulin sensitivity and muscle mitochondrial adaptation produced by exercise training.[6] Participants taking it got less out of the same workouts.
For someone with type 2 diabetes that trade is usually worth making. For a metabolically healthy person taking it speculatively while also training for healthspan reasons, it is a direct conflict between two parts of the same protocol, and it applies at ordinary doses.
Dose and timing around training
One physician's account shows how this plays out in practice. Peter Attia wrote in 2019 that while on metformin he saw slightly higher blood lactate during steady low-intensity aerobic sessions, read it as a possible sign of reduced mitochondrial efficiency, then lowered his dose and moved it to the evening so less drug was circulating while he exercised.
How it fits the trial result
That is one person's self-measurement, not a trial, but it matches the direction of the controlled result above.[6] For anyone asking how to take metformin for longevity while training, dose and timing are the two levers worth raising with a prescriber, rather than a reason to self-adjust.
- What this does not say:
- A self-report of his own lactate readings, not a controlled measurement; lactate at a given effort varies with fitness, diet and sleep.
If you are young and a bodybuilder and the muscle is important for you, don't take metformin.
What limits how much metformin a person can take?
Two things set the limit: kidney function sets the ceiling, and the digestive tract sets the practical limit below it.
Kidney function sets the ceiling
Kidney function, without qualification. Metformin is cleared renally, and reduced kidney function is either a contraindication or a dose limit depending on severity, because accumulation is what drives the rare serious risk of lactic acidosis. Prescribing guidance is explicit about checking this before and during treatment.
The gut sets the practical limit
Below that ceiling, the practical limit is your digestive tract. Gastrointestinal effects are the usual reason people stop, which is why titration and extended-release formulations exist and why taking it with meals is standard advice. A prescriber sets the metformin dosage for longevity use against both limits, not against an aging outcome.
Which long-term metformin effect goes unnoticed without testing?
Vitamin B12 depletion is the best-documented chronic effect of long-term use, it develops silently, and it is found only if someone checks. Anyone taking metformin indefinitely, particularly on their own initiative rather than for diabetes, should have B12 on the monitoring panel alongside kidney function and HbA1c.[7]
What each test on the panel is for
- Kidney function: checked before the first prescription and periodically after, because it decides whether metformin can be used at all and at what ceiling.
- Vitamin B12: the slow, silent deficit of long-term use, which no symptom flags early.
- HbA1c: a glucose marker, useful context, but not a measure of aging.
Notice what is missing. There is no test that shows whether a given dose is slowing aging, which is the plainest way to see why every quoted number stays borrowed.
Does microdosing metformin have any evidence?
None. No trial has tested forum schedules, such as a few hundred milligrams on alternate days or only on rest days from training, against an aging outcome. The logic is intuitive: keep some benefit, shed the side effects and the exercise penalty.
But no benefit has been shown at full doses to preserve at lower ones. A forum dose is a guess layered on a borrowed number, so the same kidney and B12 questions still apply to anyone following one.
How should you read a metformin dosage recommendation?
Ask what indication the number came from. A diabetes dose is doing a job unrelated to the reason you are considering it, and a longevity clinic's dose came from diabetes practice via a prescriber with no outcome data either. Rapamycin dosing for longevity faces a similar gap between protocol and outcome.
Where the dose question leads
The dose question sits downstream of a larger one. Our page on what the metformin longevity evidence actually shows covers that, and the comparison with rapamycin explains why the two are no longer in the same evidential position. If you are training, the creatine page covers a cheaper intervention that helps training rather than blunting it.
Frequently asked questions
What is the metformin dosage for longevity?
There is no approved or validated one. The trial designed to test the question proposes 1,500 mg daily over six years, and clinics prescribing it off-label commonly use 500 mg to 2,000 mg daily. Those are protocol figures and prescribing conventions, not results.
Does taking metformin in the evening avoid the exercise penalty?
Nobody has tested that. One physician described moving his dose to the evening so less drug circulated while he trained, which is a self-report rather than a trial. Timing is a reasonable question to raise with a prescriber, not something the evidence has settled.
How do doctors start metformin for anti-aging?
Typically low and slow: 500 mg daily or several times a week, increasing over weeks to reduce gastrointestinal upset. That titration manages tolerability. It does not identify an effective dose, because none has been identified.
Is 500 mg of metformin enough for longevity?
Enough for what is the missing half of the question. Doses at the lower end reduce gastrointestinal effects and no dose has been shown to slow aging in a healthy person, so there is nothing to be enough for yet.
Does metformin dosage depend on kidney function?
Absolutely, and this is the most clinically important part of any metformin conversation. Reduced kidney function is a contraindication or a dose limit depending on severity, because metformin is cleared renally and accumulation is what drives the rare serious risk.
Can healthy people take metformin?
Only by prescription and off-label. Diabetes journals have argued explicitly against extending metformin into prediabetes on the basis of the underlying risk profile, which is a more conservative position than most longevity clinics take.
Does metformin blunt exercise gains at any dose?
A randomized trial in older adults found metformin attenuated the improvement in insulin sensitivity and muscle mitochondrial adaptation from exercise training. That is a real trade-off and it applies at ordinary therapeutic doses.
Should metformin be taken with food?
Yes, with meals, which reduces the gastrointestinal effects that cause most people to stop. Extended-release formulations are also better tolerated than immediate-release at equivalent totals.
What monitoring goes with metformin?
Kidney function before starting and periodically after, and vitamin B12 over the longer term, since depletion is the best-documented chronic effect and goes unnoticed unless someone checks. HbA1c is usually monitored too.
Is 3,000 mg of metformin safe?
That is above the usual maximum in diabetes practice and would be an unusually high dose. Nothing about the longevity hypothesis argues for exceeding standard therapeutic amounts, and the risk profile worsens as the dose rises.
Did metformin extend lifespan in animals?
Not reliably, and this is the awkward center of the whole topic. A 2025 review noted that metformin given to mice from mid-life failed to produce significant lifespan benefits, and a large 2025 meta-analysis across vertebrates found no consistent effect.
Does the dose matter if the effect is unproven?
It matters for risk. Gastrointestinal effects, B12 depletion and the kidney limit all scale with exposure, so someone taking it speculatively is accumulating those at whatever dose they choose while the benefit remains hypothetical.
What about microdosing metformin?
No trial has tested sub-therapeutic dosing for any aging endpoint. It is a reasonable-sounding way to reduce side effects and there is no evidence that it retains any benefit, since no benefit has been established at full doses either.
What is the honest position on metformin dosage?
Every number in circulation is borrowed from diabetes practice. If you take it off-label, the practical questions are kidney function, B12 monitoring, and whether you are also training, because the exercise interaction is the most concrete downside anyone has measured.
References
7 sources, all link-checked; oldest check
- Metformin as anti-aging therapy: is it for everyone?. Soukas AA, Hao H, Wu L. Trends in Endocrinology and Metabolism 2019 (cited 314).Review of the anti-aging case and its limits, including the question of who might not benefit.Verified Sep 15, 2026
- Emerging uncertainty on the anti-aging potential of metformin. Keys MT et al. 2025 (cited 20).Recent review re-examining the metformin longevity case, including the 5.8% mean lifespan increase reported in model organisms and the confounding in the human observational data.Publisher blocks automated checks · last tried Sep 15, 2026
- Metformin as a tool to target aging. Barzilai N et al. Cell Metabolism 2016;23(6):1060–1065.Rationale for the TAME (Targeting Aging with Metformin) trial.Publisher blocks automated checks · last tried Sep 15, 2026
- Metformin inhibits mitochondrial adaptations to aerobic exercise training in older adults. Konopka AR et al. Aging Cell 2019;18(1):e12880.Publisher blocks automated checks · last tried Sep 15, 2026
- Metformin decelerates biomarkers of aging clocks. Abou Zaki R et al. 2024 (cited 28).In treated monkeys, protein age decreased by an average of 6.41 years across tissues showing signs of aging.Verified Sep 15, 2026
- TAME: Targeting Aging with Metformin. American Federation for Aging Research.The proposed multi-center trial designed to test whether metformin delays age-related disease; as of 2026 it has published no efficacy results.Publisher blocks automated checks · last tried Sep 15, 2026
- Rapamycin, not metformin, mirrors dietary restriction-driven lifespan extension in vertebrates. Ivimey-Cook ER et al. Aging Cell 2025;e70131 (cited 10).Meta-analysis pooling 167 studies across vertebrate species; dietary restriction and rapamycin extended lifespan consistently, metformin did not.Publisher blocks automated checks · last tried Sep 15, 2026
Every link above is re-requested on a schedule by an automated checker; the date shown is when it last answered. Publishers that block automated requests are marked as such rather than reported broken. Our editorial standards explain how a source gets cited here in the first place.
Filed by the Rapamycin.store evidence desk. Each source is re-checked when the page is reviewed.
Own a longevity or health brand?
Rapamycin.store publishes sponsored and guest posts on an aged, topically-relevant domain, and offers display advertising. Tell us what you want to promote and we'll send our current options.