Buy GitHub stars cheap: what the lowest prices actually buy

September 23, 2026 · 11 min read

Type "buy GitHub stars cheap" into a search engine and the prices look impossible. A thousand stars for $9. Five stars for $2.50. A "premium" listing asking $90 for fifty. Every one of them is advertised as real, and none of them is the same product.

The cheap end of this market is real - and it is where almost all of the risk lives. This page puts the live price board next to the math the price tags hide, so you can tell the difference between a bargain and a bill that arrives later.

The price board, September 2026

Star prices are usually quoted per unit, which makes them hard to compare. The same listings per thousand:

Listed at	Rate	Per 1,000	What the listing says
Follownic	$0.01/star, $13.50 per 1,000	$13.50	"No refill" and 30-day refill options sold side by side
Promobox	$0.009/star worldwide	$9	"Real Users" label; Russia-targeted costs 4x more
SMGrowth	$25 for 500	$50	180-day refill guarantee, gradual processing
GithubStarMate	$5 for 20 actions	$250	Each star, watch or fork spends one credit
Fiverr gigs	$0.45/star (reported May 2026)	$450	Marketplace listings; seller ratings, no platform verification of accounts
SocialPanda / SocialPlug	$0.201/star	$201	"Stars from real people", up to 1,000 stars
NiubiStar	¥1.8/star (~$0.25)	~$250	Manual pacing, 24-120 hours, Alipay
GetAFollower	$2.50 for 5 stars	$500	Entry package; monthly plan offers 10% off
Upvote.club	from $0.90/star	$900	Community requests - members choose whether to participate
Twidium	$90 for 50 stars	$1,800	Manual, 6-30 day delivery, old-school pricing
RepoBoost (this site)	$30 per 1,000 actions	$30	Stars + forks + watches + followers, aged accounts, no-drop

The board is the point. "Cheap" is not a price band in this market - it is a 200x range that hides three different products: bulk fresh-account stars, per-unit mid-market delivery, and small-batch community or manual work. Two listings from the same search results can differ by 55x per star.

Why $9 per 1,000 exists

A star costs one click. The only input a vendor needs is an account willing to click, and freshly registered accounts cost almost nothing to produce in bulk. That is the entire economics of the cheapest tier: empty profiles, created weeks or days before use, executing an order in one burst.

That pattern has a name in the research. The ICSE 2026 fake-star study classifies it as the low-activity signature - accounts with no repositories, no followers and no activity history, starring in synchronized batches. The study's detector flagged campaigns of exactly this shape across 18,617 repositories and roughly 301,000 accounts.

Notice what the cheap listings do not include. "No refill" options are sold openly next to refill options, which tells you the vendor expects drops and prices the risk accordingly. The $9 thousand sits on the same page as a 30-day refill add-on because without it, the stars are a rental.

How it got this cheap

The floor did not fall because stars became easier to deliver well. It fell because accounts became cheaper to mass-produce - and the market has receipts. In March 2023, the Dagster team bought 1,000 stars for $64 (about $0.064 each) from a bulk vendor and reported that roughly three quarters had survived. Three months later, The Guild paid €8.19 for 100 stars from a budget seller; the accounts had been created the same week as the order, and a month later all of them were gone. The same investigation paid €0.80 per star to a premium vendor whose accounts looked human - and those survived. WIRED's test purchase, $6 for 50 stars, was removed within weeks.

Read the three experiments together and the structure is obvious. The cheap tier keeps collapsing - from $64 per 1,000 in 2023 to $9 per 1,000 in 2026 - because it sells disposable accounts. The mid tier has barely moved in three years, because it prices delivery infrastructure rather than account supply. The premium tier never got cheap, because history cannot be mass-produced. When you compare two quotes, you are not comparing margins. You are comparing which of those three things you are buying.

The math: cost per surviving star

Sticker price is the wrong number to compare. The number that matters is what you paid per star that is still there next quarter - including replacements.

Run the arithmetic on a cheap order with no refill terms:

• You pay $13.50 for 1,000 stars.
• A cleanup wave removes 40% - a plausible rate for fresh-account pools, and the exact failure mode the research documents.
• 600 stars remain. Your cost per surviving star is now $22.50 per 1,000.
• You rebuy the 400 lost stars at the same rate - another $5.40 - and the original thousand has cost $18.90, with no forks, no watches, no followers, and no guarantee the replacements survive either.

Compare that against a flat $30 for 1,000 actions that arrive together - 700 stars plus 75 forks, 200 followers and 25 watches - paced over 12-24 hours from aged accounts, with anything that drops replaced free. The $9 sticker can end up more expensive than the $30 order while delivering less, and that is before the research finding that fake stars have a promotion effect about 5x smaller than real stars and turn negative after roughly two months.

A discount on a signal that decays is not a discount. It is a deferred cost.

Red flags in a cheap offer

1. It asks for credentials. A password, a token, an OAuth grant, collaborator access - none are needed to star a public repository. Anyone asking is creating a security problem you did not order.
2. Delivery is instant. A thousand stars in minutes is a burst, and the burst is the signature. Pacing is not a premium feature; it is the difference between an audience and an order.
3. No written replacement terms. "We will look into it" is not a policy. If drops are not replaced in writing, the risk has been transferred to you.
4. Per-star pricing with custom-size surcharges. Compare the effective rate for your exact order, not the headline tier. Some services also charge more for mixes or exact numbers; the price should not depend on the shape of your order.
5. Promises about ranking, trending or traffic. No vendor controls GitHub's algorithms. Ranking is a separate discipline, and any star seller bundling those guarantees is selling outcomes it does not own.
6. No delivery window and no sample. If the listing will not say when the stars arrive or what the accounts look like, there is nothing to verify after payment.

Marketing, not misrepresentation

There is a reason this market attracts regulators as well as vendors. The FTC's Consumer Review Rule, effective October 2024, prohibits buying or selling fake indicators of social-media influence, with penalties up to $53,088 per violation - and by December 2025 the agency had sent warning letters to ten companies. Securities law applies separately when inflated metrics are used to raise money, and the incentive is documented: Redpoint's data puts the median seed-stage repository at 2,850 stars, a threshold that 2026's cheapest rates bring within a few hundred dollars of any budget.

None of that makes paid visibility illegal. It draws a line. Buying distribution for a real project is marketing; presenting purchased engagement as evidence of organic traction to investors, acquirers or customers is misrepresentation, and that decision belongs to the buyer, not the vendor. Keep the two separate and the compliance risk stays theoretical - and the number keeps doing the one job it can honestly do.

The actually cheapest first move

Before paying anyone anything, run the free route. Most repositories that "need" stars have a first-screen problem: the README does not explain the project in five seconds, there is no demo, no release, no clear install path. Fixing that costs zero and converts better than any purchased number.

Start with the free launch checklist - ten readiness checks scored 0-10 - then follow the first-100-stars guide. When you do buy, track the curve yourself with the open-source star history tracker; it reads GitHub's public star-history endpoint and shows weekly and daily counts. A slope you can verify beats a discount you cannot.

Where RepoBoost sits on the board

We are the $30 row, and there is a reason the price is flat: it removes the per-star games. One thousand actions for $30, balanced packages at $30, $40 and $50, never above $50 per order. Every action comes from an aged account with real commit history - the 2009-2024 inventory we sell separately on the aged accounts page. Delivery is paced over 12-24 hours, drops are replaced free, and exact numbers or custom mixes cost the same as presets.

That is the cheap package and the expensive package at once. If you want to see how the market got its price spread, the pricing guide covers the tiers; if you want the risk side in full, the safety guide covers enforcement data. For the delivery mechanics themselves, start with the delivery method.

FAQ

How cheap are GitHub stars?

Publicly listed prices in September 2026 range from $0.009 to $1.80 per star - a 200x spread. The cheapest listings are stars-only from fresh accounts with weak replacement terms. Serious flat-rate delivery runs about $0.03 per action, including forks, watches and followers.

Why are some GitHub stars so cheap?

Because fresh accounts cost almost nothing to create. Empty profiles registered in bulk can star repositories immediately, with no history to maintain and no replacement promise to honor. That is the pattern the ICSE 2026 study calls the low-activity signature.

Is the cheapest option risky?

The research says yes for the repository, not for the buyer's account. In the ICSE 2026 study, 90.42% of repositories flagged for fake-star campaigns were eventually deleted by GitHub, while the vendor who sold the stars kept nothing at risk. Cheap is only cheap if the stars survive.

What is cost per surviving star?

Sticker price divided by the share that stays - including replacements. A $13.50 order that loses 40% to a cleanup costs $22.50 per 1,000 surviving stars; rebuying the lost stars pushes it higher, without adding forks, watches or followers. This is the only comparison that means anything.

Can I buy stars without per-star pricing?

Yes - RepoBoost charges a flat $30 per 1,000 actions, with balanced packages at $30, $40 and $50, never above $50 per order. Exact numbers and custom mixes cost the same, and every order includes stars, forks, watches and followers together.

Does RepoBoost have a cheap package?

The Starter package is $30 for 1,000 actions: 700 stars plus 75 forks, 200 followers and 25 watches, delivered from aged accounts over 12-24 hours with a no-drop guarantee. It is the lowest price we offer - and the highest, since packages never exceed $50.

Sources

• He, Yang, Burckhardt, Kapravelos, Vasilescu, Kästner. Six Million (Suspected) Fake Stars on GitHub. ICSE 2026. DOI 10.1145/3744916.3764531.
• Dagster. Detecting Fake GitHub Stars with Dagster. March 2023 - source of the budget/premium price tiers.
• 36Kr (EU edition). GitHub Stars Priced at 0.5 Yuan Each: AI Projects Most Aggressive in Buying Fake Stars. April 2026.
• The Guild. How Much Are GitHub Stars Worth to You? June 2023 - a controlled buy-and-observe test of cheap versus premium stars.
• WIRED. The GitHub Black Market That Helps Coders Cheat the Popularity Contest. October 2023.
• Federal Trade Commission. Final rule banning fake reviews and testimonials. August 2024; warning letters to ten companies reported April 2026.
• Market snapshot: publicly listed rates captured 23 September 2026 from Follownic, Promobox, SMGrowth, GithubStarMate, SocialPanda, SocialPlug, NiubiStar, GetAFollower, Upvote.club and Twidium product pages. Prices change; verify current listings before ordering anywhere.