DEV Community

Dev Jhawar profile picture

Dev Jhawar

404 bio not found

Joined Joined on 
Bigger AI models aren't always better: how a 70B model beat a 280B one by reading more

Bigger AI models aren't always better: how a 70B model beat a 280B one by reading more

Comments
6 min read
What it actually costs to train an LLM: GPUs, days and dollars, worked out step by step

What it actually costs to train an LLM: GPUs, days and dollars, worked out step by step

Comments
6 min read
Why Meta trained a small 8B model on 15 trillion tokens, and why that's cheaper than it sounds

Why Meta trained a small 8B model on 15 trillion tokens, and why that's cheaper than it sounds

Comments 1
5 min read
The KV cache is eating your GPU. MQA and GQA are how models fight back

The KV cache is eating your GPU. MQA and GQA are how models fight back

1
Comments 1
3 min read
Tokenization is why your agent bill looks weird

Tokenization is why your agent bill looks weird

Comments
4 min read
Why your agent gets slower and pricier with every tool call

Why your agent gets slower and pricier with every tool call

Comments 2
5 min read
Running 100 agents on a free tier (Groq 30 RPM, Tavily credits)

Running 100 agents on a free tier (Groq 30 RPM, Tavily credits)

Comments
4 min read
loading...