Opportunities › Weekly GPU compute marketplace with liquidity › Proposed app
Weeklot
Buy, resell and price-lock GPU nodes by the calendar week
- Score
- 37
- confidence 53
- Evidence
- 6
- items on the page
- Rivals
- 2
- named in sources
- First version
- 55 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2006
- Mature market · AI estimate
ScreensConcept
Desktop app: ML teams make purchasing decisions and manage GPU reservations in dense pricing and inventory tables at their workstations. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
H100 8x market
Sample data$2.41/GPU-hr
About this appAI proposal
Weeklot is an exchange for GPU nodes sold by the calendar week. ML teams buy only the weeks they need, see the posted clearing price before they commit, and list unused weeks for resale if plans change.
Sellers such as providers, labs with spare capacity and teams holding leases post nodes and weeks at a price. Buyers see one public order book per GPU type instead of negotiating privately, so a price spread between buyers is visible.
Teams can also reserve future weeks at today's price to hedge cost swings. Weeklot does not replace the underlying provider; it handles listing, matching, settlement and handover of access details.
- Tagline
- GPU compute by the week. Buy it, sell it, lock it.
- Voice
- PlainPreciseCalm
- Why this name
- Week plus lot: the unit of trade is a lot of compute for one calendar week. Short, pronounceable, easy to say aloud.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Week-sized purchases
Buy a node for one or more calendar weeks instead of a 6-24 month lease.
No member yet
Public clearing prices
See recent clearing prices and open bids and asks per GPU type, so you know what a week costs.
No member yet
Sell back unused weeks
List weeks you will not use at a posted price and recover part of the cost instead of paying for idle nodes.
No member yet
Forward reservations
Reserve future weeks at a fixed price to plan budgets when rates swing.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#0F3B3FDeep Teal · primary#F6F4EEChalk · surface#14201FInk · text#E8A317Signal Amber · accent#2E8B6AFern · data up#B5412FBrick · data down
Typography
Aa
Weeklot
PricingAI proposal
The material gives no competitor prices. The 2-5% transaction take comes from the opportunity's monetization idea; 3% is a starting point to test.
Browse
Free
- Public clearing prices
- Order book view
- One price alert
Trader
Free
- Buy, sell and reserve weeks
- Unlimited price alerts
- Fee of 3% per transaction
Team
Contact us
- Shared team account
- Volume fee terms
- Invoicing and exports
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- Shared outputs. Weeklot publishes public price pages and embeddable price charts per GPU type. Teams cite and share them in posts, budgets and discussions, and each links back to the order book.
- Do they become customers?
- Readers of price pages are ML buyers and could trade. Test whether visitors to the public pages register and complete a first purchase.
- First customers
- Small ML startups and research labs with 8-64 GPUs needing bursty training runs, plus one or two providers with idle nodes to seed supply.
- Where to reach them
- Hacker News and ML forums: Post public price data and a short write-up of the weekly-lot idea; invite comments and sign-ups.
- Direct outreach to providers: Offer sellers with idle weeks a free listing period to fill the first order books.
- ML community Slack and Discords: Share weekly price summaries and take requests for node types.
- First milestone
- Ten completed weekly trades between at least three sellers and five buyers on one GPU type.
Why people already pay for this
- AI inferenceGPU capacity buyers need the ability to exit or redeploy unused weeks when plans change.
- VerifiedHacker News post received 46 points and 29 comments
- AI inferenceCompute infrastructure trades through private bilateral leases with no visible price, similar to energy markets before deregulation.
- AI inferenceThe same H100 GPU rents at a 2x price spread depending on buyer, and committed leases cannot be resold.
- Public claim
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Order matchingMatching engine on Postgres with transactional locksBids, asks and week lots need consistent matching without double sales.
- This app · Escrow and payoutsStripe Connect with held fundsSellers need payouts and buyers need payment held until delivery.
RisksAI inference
- CommercialTwo-sided liquidity: without enough sellers posting nodes, buyers find empty books and leave. Supply must be secured before demand.
- PlatformMany leases and provider terms may forbid resale or sublicensing, so listed weeks could breach seller contracts.
- RegulatoryForward price reservations may be treated as financial derivatives in some jurisdictions and need legal review.
- TechnicalDelivering access to a node for one week and proving it worked is hard; disputes over performance need a clear process.
- DataEarly clearing prices come from few trades and may be thin or easy to move, which weakens the transparency claim.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.