The cheapest documented offers, priced per unit

"Cheapest" is the most abused word in this market, and it is why this page exists. The cheapest headline price per TH, per GPU-hour, or per TB is a marketing number; the cheapest contract is the one with the lowest all-in cost once maintenance, term, payout threshold, and exit friction are included. The ranking separates the two, and this page explains how - including the three corners where cheap is an honest signal.

Headline versus effective price

Work an example at today's numbers: BTC hashprice sits at n/a per TH/day. A contract priced at a $0.05/TH/day maintenance fee starts underwater at any headline price - the fee alone exceeds gross revenue [1]. Meanwhile a listing at a higher headline with a zero fee can be the cheaper deal over a year. The gap between the two bars in the figure is the actual cost of "cheap."

The mining calculator prices this arithmetic per provider: up-front price, maintenance, term, and live hashprice against a difficulty-growth assumption you control.

Where cheap is honest

Cheap is a real signal in exactly three corners of this market:

• Spot hashrate markets - day-to-day rental priced near hashprice with a tight spread. The option to walk away is the product; the cheap rate is honest because it floats [1].
• Community GPU marketplaces - consumer GPUs at commodity rates. The rate is real; the discount is interruption risk, which your job design has to absorb [3]. The current 3090 floor in our feed is cheap for exactly this reason.
• Off-peak compute bids (Akash-style) - bidding below the clearing price gets you capacity that would otherwise idle [4]. The discount is real; the interruptability is the price.
• Storage at write-heavy ratios - if you never read the data, the retrieval fee is irrelevant and headline $/TB decides. Read-heavy workloads invert the ranking (see the storage hub).

Live from our database - caveats and sources in each linked review.

Where cheap is a trap

• Long-term contracts at a discount - the discount is a forecast: the operator priced a difficulty path you cannot see. The cheap headline is how they sell the forecast error to you [1].
• Altcoin contracts with no pool transparency - thin-liquidity coins are where fake hashrate pricing hides. Verify the pool worker before accumulating. The hashrate markets hub explains how FPPS-versus-PPLNS mechanics shift the same warning onto market rentals.
• Lifetime contracts - the cheapest per-month number in the table is usually a claim that outlives the company. The scam watch archive is the ledger of how that resolves.
• "Free" electricity or free-hashrate offers - power costs break even near n/a per kWh today; anyone running machines for free is recovering the cost elsewhere [2].

A worked comparison at today's numbers

Take two offers for the same 100 TH and price them honestly [1]:

• Offer A: $99 headline for 360 days, plus a $0.0332/day service fee deducted from mined BTC. Gross revenue at n/a per TH/day is $0.39/day; the fee takes 8.5% of it. Over the term the fee takes roughly $12 against $143 gross - the effective cost is the up-front price plus $12 in lost revenue.
• Offer B: a market rental at 10% over hashprice with zero term lock. Same gross revenue, 10% spread, exit any day. If hashprice falls 30% in month two, A keeps billing and B walks away.

Neither number alone decides the comparison - the difficulty path and your holding period do. That is precisely what the calculator and the contract vs HODL tool exist to price. Run both before any purchase, because the honest answer changes with the trend: our feed's current BTC movement is n/a, and that single number can flip a marginal decision either way.

The honest cheapest framework

Rank by all-in cost per unit of expected output at your workload, not by headline:

1. Price the term: up-front + maintenance over term.
2. Price the exit: payout threshold, withdrawal cadence, and what an unclaimed balance is worth at term end.
3. Price the risk: operator age, custody model, proof level - the columns the providers table carries for every row.

A contract that wins on step 1 and loses on step 3 is not cheap. This is the entire argument for the contract vs HODL tool: the comparison is cash flow, not rate cards. The current population spans 83 live providers with entry prices from $0.01 - the spread between the cheapest headline and the cheapest effective deal is the single most profitable piece of arithmetic on this page.