Top all-time · #18766
Radiant Blog
Company blog · blog.ori.coaiprogramming
Radiant Blog is the company blog at blog.ori.co, writing about ai and programming. Since 2023, 4 different Hacker News readers have linked it 5 times. No feed was found, so read it on the site itself.
Too few sampled comments (1) to say how often HN recommends it.?
Why HN links it
The Radiant blog covers AI infrastructure and GPU compute economics. The one HN comment compares GPU-hour training costs of Llama 2 and JetMoE, without judging the posts themselves.
Summarised by Claude from a sample of 1 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Explore technical posts, analysis and commentary from Radiant on AI infrastructure, energy, and the age of abundance powering the next industrial revolution.
What Hacker News says?
It looks like Llama 2 7B took 184,320 A100-80GB GPU-hours to train[1]. This one says it used a 96×H100 GPU cluster for 2 weeks, for 32,256 hours. That's 17.5% of the number of hours, but H100s are faster than A100s [2] and FP16/bfloat16 performance is ~3x better. If they had tried to replicate Llama 2 identically with their hardware setup, it'd cost a little bit less than twice their MoE model. [1] https://github.co…
Comment linking /choosing-between-nvidia-h100-vs-a100-performance-and-costs-considerations
on “JetMoE: Reaching LLaMA2 performance with 0.1M dollars”Submitted Kubernetes Made Easy for AI as a story · 2 points
HN discussionSubmitted Hands-On Change Data Capture in Go with Postgres and Debezium as a story · 2 points
HN discussionSubmitted Hands-On Change Data Capture in Go with Postgres and Debezium as a story · 1 points
HN discussionSubmitted Multi Cloud for AI Companies as a story · 1 points
HN discussion