Computable GPU Index (CGI) is a USD price per GPU-hour, computed from the published on-demand rental rates of a fixed panel of providers. The methodology is mathematically robust, and anyone can verify and reproduce every print.
Hi Product Hunt, I'm Ray, cofounder of Computable.
What is the price of a GPU?
Nobody agrees. Every provider quotes a different number, and existing indexes are closed black boxes: a figure you are asked to trust without seeing the data, the method, or the code behind it. Despite being rented, resold, and financed at commodity scale, compute lacks the reference rate every mature commodity market has.
CGI is built to be that number, and to earn trust in the open rather than by authority. It is open data, open method, open code.
How it works: every 15 minutes we collect published on-demand rental rates from 28 providers. Each panel provider casts weighted votes at its price, and the index is an interquantile mean over those votes, not an average: only the central third of the vote mass is averaged. No small group of providers can drag the number from the tails.
The weights are systematically scored. Liveness weights measure whether a provider contributes new information to the price: a leave-one-out ridge regression scores whether each provider's recent moves anticipated the rest of the panel, and the weights recompute every observation. Every weight is floored and capped, and no weight requires a per-provider human decision.
The design rests on five properties: verifiable, reproducible, fault-tolerant, outlier-resistant, transparent. The data is verifiable, the method is transparent, the prints are reproducible via code, and the result is a robust price surviving data sources that disagree, err, or break.
Live today: H100, H200, B200, B300. The collector and the calculation are open source. Clone the repo and recompute any print since inception; you'll get the same number we published.
If you trade, rent, or finance compute, I'd love to hear what you want from a reference rate.
Report
@raysong nice launch congrats🙌awesome to see a fault-tolerant approach to this
@raysong@priya_kushwaha1 Totally agree, the fault tolerant design feels especially important for something aiming to become a trusted market benchmark.
@raysong The transparency here is what caught my attention. Publishing the methodology, source data, and code makes CGI much easier to scrutinize than a typical closed index. I upvoted Computable. I’m curious how you plan to handle providers changing pricing models or introducing new GPU configurations over time.
@natalia_iankovych compute is a very inefficient market with a lot of local price fluctuation. we have built the price index with that in mind and made sure that the index is resistant to outliers.
Congrats for your launch! I'm really interested in your CGI! Also, I have a tiny question: Could CGI eventually help teams compare the cost of training across GPU generations?
@brattyaiguy Thank you! And yes, that's a big part of exactly what CGI is for. It gives you the price leg: a comparable $/GPU-hour for each generation. they are under one methodology, so the prices are apples to apples by construction.
however, you will still have evaluate the performance. We deliberately don't bake performance assumptions into the index itself, since every workload scales differently, but pairing your benchmarks with a clean price series from our index is a great way to make comparisons.
Computable GPU Index (CGI)
Hi Product Hunt, I'm Ray, cofounder of Computable.
What is the price of a GPU?
Nobody agrees. Every provider quotes a different number, and existing indexes are closed black boxes: a figure you are asked to trust without seeing the data, the method, or the code behind it. Despite being rented, resold, and financed at commodity scale, compute lacks the reference rate every mature commodity market has.
CGI is built to be that number, and to earn trust in the open rather than by authority. It is open data, open method, open code.
How it works: every 15 minutes we collect published on-demand rental rates from 28 providers. Each panel provider casts weighted votes at its price, and the index is an interquantile mean over those votes, not an average: only the central third of the vote mass is averaged. No small group of providers can drag the number from the tails.
The weights are systematically scored. Liveness weights measure whether a provider contributes new information to the price: a leave-one-out ridge regression scores whether each provider's recent moves anticipated the rest of the panel, and the weights recompute every observation. Every weight is floored and capped, and no weight requires a per-provider human decision.
The design rests on five properties: verifiable, reproducible, fault-tolerant, outlier-resistant, transparent. The data is verifiable, the method is transparent, the prints are reproducible via code, and the result is a robust price surviving data sources that disagree, err, or break.
Live today: H100, H200, B200, B300. The collector and the calculation are open source. Clone the repo and recompute any print since inception; you'll get the same number we published.
If you trade, rent, or finance compute, I'd love to hear what you want from a reference rate.
@raysong nice launch congrats🙌awesome to see a fault-tolerant approach to this
Acti
@raysong @priya_kushwaha1 Totally agree, the fault tolerant design feels especially important for something aiming to become a trusted market benchmark.
Computable GPU Index (CGI)
@priya_kushwaha1 thank you!
@raysong Congrats on the launch!
Computable GPU Index (CGI)
@oussama_nakhil thanks!
@raysong The transparency here is what caught my attention. Publishing the methodology, source data, and code makes CGI much easier to scrutinize than a typical closed index. I upvoted Computable. I’m curious how you plan to handle providers changing pricing models or introducing new GPU configurations over time.
Tiny CV
Congrats on launching a super cool product. Awesome that you guys launched with open API access and an MCP!
Computable GPU Index (CGI)
@andrewjiang thank you! glad you liked it!
One of the providers stands out quite a lot. Is there a mistake here?
Computable GPU Index (CGI)
@natalia_iankovych compute is a very inefficient market with a lot of local price fluctuation. we have built the price index with that in mind and made sure that the index is resistant to outliers.
Memmy Agent
Congrats for your launch! I'm really interested in your CGI! Also, I have a tiny question: Could CGI eventually help teams compare the cost of training across GPU generations?
Computable GPU Index (CGI)
@brattyaiguy Thank you! And yes, that's a big part of exactly what CGI is for. It gives you the price leg: a comparable $/GPU-hour for each generation. they are under one methodology, so the prices are apples to apples by construction.
however, you will still have evaluate the performance. We deliberately don't bake performance assumptions into the index itself, since every workload scales differently, but pairing your benchmarks with a clean price series from our index is a great way to make comparisons.
PageIndex
Congrats on the launch! Do you include hyperscaler list prices in the provider panel?
Computable GPU Index (CGI)
@cathy_cc we do not. hyperscaler prices are not always available, because different players can get different terms and guarantees.
Acti
Would you consider showing a source-coverage indicator alongside each index value?
Computable GPU Index (CGI)
@axelkane hey it's already there. you can see it on the bottom left of the dashboard.
Acti
How frequently are provider prices collected?
Computable GPU Index (CGI)
@mati_lee every 15 minutes