AURORA INSIGHTS
AURORA INSIGHTS
Check back for new articles published every Monday at 9 AM UTC.
DeepSeek: Smaller, Cheaper, Better?
With the cost of over $100 Million for training GPT-4, not many enterprises can compete at this budget level. But what if there was a model that cost under $6 Million to train that matches GPT-4o’s performance? Enter: DeepSeek.
Pricing LLMS: The True Cost of Compute
After browsing through the technical paper of Llama 3.1, I came to the conclusion that LLMs are “Larege Luxury Models.” The cost to train these models has increased exponentially over the years and the performance seems to be peaking, it raises the question: is there a more efficient way forward for LLM innovation?