DeepSeek Outpaced OpenAI at 3% of the Cost
(venturebeat.com)
DeepSeek R1’s Monday release has sent shockwaves through the AI community, disrupting assumptions about what’s required to achieve cutting-edge AI performance.
DeepSeek R1’s Monday release has sent shockwaves through the AI community, disrupting assumptions about what’s required to achieve cutting-edge AI performance.
Ask HN: Cheapest way to run a time-series database in cloud?
(ycombinator.com)
I need to run a database (preferably Postgres-based, currently using TimescaleDB) to store about 20M rows of price data per day, with the option to discard or offload data after 7d to cold storage.
I need to run a database (preferably Postgres-based, currently using TimescaleDB) to store about 20M rows of price data per day, with the option to discard or offload data after 7d to cold storage.
How we saved millions on AWS
(forter.com)
In early 2022, as the world was emerging from the Covid-19 pandemic, inflation surged to multi-decade highs, prompting central banks to raise interest rates.
In early 2022, as the world was emerging from the Covid-19 pandemic, inflation surged to multi-decade highs, prompting central banks to raise interest rates.
PR Previews Don't Need Vercel: My Solution on a $5 VPS
(pert.dev)
Let's preface this by admitting that the title is a little farcical - the fact that Vercel and other similar sites give you preview domains out of the box with no setup is a lovely feature. This blog is purely to show you that it's actually possible to configure these yourself, and furthermore, it's way easier than you might think!
Let's preface this by admitting that the title is a little farcical - the fact that Vercel and other similar sites give you preview domains out of the box with no setup is a lovely feature. This blog is purely to show you that it's actually possible to configure these yourself, and furthermore, it's way easier than you might think!
Using AZs can eat up your budget – From Prometheus to VictoriaMetrics
(prezi.com)
By 2024, Prezi’s monitoring system, built around Prometheus, was becoming outdated. It was already 5+ years old, running on a deprecated internal platform and accumulating a significant amount of costs every month.
By 2024, Prezi’s monitoring system, built around Prometheus, was becoming outdated. It was already 5+ years old, running on a deprecated internal platform and accumulating a significant amount of costs every month.
Scalable Server SQLite Apps
(servicestack.net)
Ever since adding support for Litestream in our project's templates GitHub Action Deployments we've been using SQLite as the backend for our new .NET C# Apps as it's the most cost-effective option that frees us from needing to use a cloud managed database which lets us make use of Hetzner's much cheaper US Cloud VMs.
Ever since adding support for Litestream in our project's templates GitHub Action Deployments we've been using SQLite as the backend for our new .NET C# Apps as it's the most cost-effective option that frees us from needing to use a cloud managed database which lets us make use of Hetzner's much cheaper US Cloud VMs.
Cutting AWS costs through inference infrastructure improvements
(vannevarlabs.com)
Vannevar Labs, a defense tech startup, successfully cut machine learning (ML) inference costs by 45% using Ray and Karpenter on Amazon Elastic Kubernetes Service (Amazon EKS).
Vannevar Labs, a defense tech startup, successfully cut machine learning (ML) inference costs by 45% using Ray and Karpenter on Amazon Elastic Kubernetes Service (Amazon EKS).
Delivering 15TB of 4K video with Cloudflare R2 for $2.18
(screencasting.com)
We serve a lot of video. In fact, just last month, our viewers watched over 15 terabytes of it—and with several new courses planned, our monthly bandwidth will more than double next year.
We serve a lot of video. In fact, just last month, our viewers watched over 15 terabytes of it—and with several new courses planned, our monthly bandwidth will more than double next year.
Show HN: Costco for LLM Tokens
(inference.net)
inference.net is a wholesaler of LLM inference tokens for models like Llama 3.1. We provide real-time and batch inference APIs at a 50-90% discount from what you would pay together.ai or groq. If you're spending > $10K/month on inference, we can likely reduce your costs substantially. You can reach us at support@inference.net.
inference.net is a wholesaler of LLM inference tokens for models like Llama 3.1. We provide real-time and batch inference APIs at a 50-90% discount from what you would pay together.ai or groq. If you're spending > $10K/month on inference, we can likely reduce your costs substantially. You can reach us at support@inference.net.
Tell HN: Switched from Lightsail to Hetzner Cloud, 2 blogs for $4 a month
(ycombinator.com)
After a few comments mentioned I was probably wasting money using AWS Lightsail, I finally tried out Hetzner Cloud for hosting my Ghost Blogs.
After a few comments mentioned I was probably wasting money using AWS Lightsail, I finally tried out Hetzner Cloud for hosting my Ghost Blogs.
Own Infrastructure Instead of AWS: Significantly Lower Costs, No Hidden Fees
(heise.de)
37signals aims to save just under two million US dollars a year by moving out of the cloud and back into its own data center – significantly more than the initially targeted seven million dollars over five years.
37signals aims to save just under two million US dollars a year by moving out of the cloud and back into its own data center – significantly more than the initially targeted seven million dollars over five years.
Video scraping: extracting JSON from a 35s screen capture for 1/10th of a cent
(simonwillison.net)
The other day I found myself needing to add up some numeric values that were scattered across twelve different emails.
The other day I found myself needing to add up some numeric values that were scattered across twelve different emails.
Audioscrape: Building in Rust When Everyone Said I Shouldn't
(ycombinator.com)
I'm excited to share my journey of bootstrapping Audioscrape, a podcast exploration platform, built entirely in Rust. Despite conventional wisdom suggesting RoR, Python, or TypeScript for rapid MVP development, I chose Rust to challenge myself technically and optimize for low operational costs. The result? A performant application running on a $7/month VM, demonstrating that you can launch lean and scale efficiently.
I'm excited to share my journey of bootstrapping Audioscrape, a podcast exploration platform, built entirely in Rust. Despite conventional wisdom suggesting RoR, Python, or TypeScript for rapid MVP development, I chose Rust to challenge myself technically and optimize for low operational costs. The result? A performant application running on a $7/month VM, demonstrating that you can launch lean and scale efficiently.
Scalable Server SQLite Apps
(servicestack.net)
Ever since adding support for Litestream in our project's templates GitHub Action Deployments we've been using SQLite as the backend for our new .NET Apps as it's the most cost-effective option that frees us from needing to use a cloud managed database which lets us make use of Hetzner's much cheaper US Cloud VMs.
Ever since adding support for Litestream in our project's templates GitHub Action Deployments we've been using SQLite as the backend for our new .NET Apps as it's the most cost-effective option that frees us from needing to use a cloud managed database which lets us make use of Hetzner's much cheaper US Cloud VMs.
Ask HN: Cheap way to run a small newsletter?
(ycombinator.com)
For a newsletter of around 1,500 subscribers mailchimp wants £34 a month.
For a newsletter of around 1,500 subscribers mailchimp wants £34 a month.
Saving $10k/Month on Analytics – Snowplow Serverless Alternative
(agondata.com)
Faced with the high costs of traditional analytics solutions, our team implemented a serverless Snowplow alternative to dramatically reduce costs. By leveraging cloud-native technologies and serverless architectures, we not only achieved significant cost savings but also gained a scalable, flexible analytics platform with full data ownership.
Faced with the high costs of traditional analytics solutions, our team implemented a serverless Snowplow alternative to dramatically reduce costs. By leveraging cloud-native technologies and serverless architectures, we not only achieved significant cost savings but also gained a scalable, flexible analytics platform with full data ownership.
Show HN: Keep Your Next Viral AI App Free for Longer with Local Embeddings
(fxn.ai)
Whether you’re a solopreneur building an intelligent search app; or you’re a series A startup finding product-market fit in enterprise knowledge management; you can be saving 60% or more on your monthly OpenAI bills by generating embeddings on your users’ devices.
Whether you’re a solopreneur building an intelligent search app; or you’re a series A startup finding product-market fit in enterprise knowledge management; you can be saving 60% or more on your monthly OpenAI bills by generating embeddings on your users’ devices.
Reclaim the Stack
(reclaim-the-stack.com)
We spent 7 months building a Kubernetes based platform to replace Heroku for our SaaS product at mynewsdesk.com. **The results were a 90% reduction in costs and a 30% improvement in performance.** We also significantly improved developer experience with reduced deploy times and faster / more accessible tooling.
We spent 7 months building a Kubernetes based platform to replace Heroku for our SaaS product at mynewsdesk.com. **The results were a 90% reduction in costs and a 30% improvement in performance.** We also significantly improved developer experience with reduced deploy times and faster / more accessible tooling.