DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Local AI Updates: llama.cpp MTP, vLLM Gemma 4 Speeds, Ollama Coder Benchmarks

Local AI Updates: llama.cpp MTP, vLLM Gemma 4 Speeds, Ollama Coder Benchmarks

Comments
3 min read
The Hidden Math Behind AI Agents: Why GPT-4o Can Be More Expensive Than Hiring a Human

The Hidden Math Behind AI Agents: Why GPT-4o Can Be More Expensive Than Hiring a Human

Comments
1 min read
"What Codex's 'sudo workaround' actually means for production agents"

"What Codex's 'sudo workaround' actually means for production agents"

Comments
5 min read
Why agents need memory that improves itself

Why agents need memory that improves itself

1
Comments
2 min read
Your RL Agent Failed a 12-Step Task. Which Step Was Wrong? (The Supervision Problem in Agentic RL)

Your RL Agent Failed a 12-Step Task. Which Step Was Wrong? (The Supervision Problem in Agentic RL)

Comments 2
5 min read
Just joined the Gemma 4 Challenge by Google AI & DEV Community!

Just joined the Gemma 4 Challenge by Google AI & DEV Community!

Comments
1 min read
A Fluent LLM Answer Is Not the Same as an Inspected Answer

A Fluent LLM Answer Is Not the Same as an Inspected Answer

1
Comments
4 min read
Prompt Engineering Is Systems Design, Not a User Skill

Prompt Engineering Is Systems Design, Not a User Skill

2
Comments 1
5 min read
Stop Using 'Skills' for Brainstorming. Build a Hook Instead. 🛠️

Stop Using 'Skills' for Brainstorming. Build a Hook Instead. 🛠️

1
Comments 1
3 min read
Are We Going to an AI-First Frameworks Era?

Are We Going to an AI-First Frameworks Era?

5
Comments 1
2 min read
Building AI agents with OpenAI Agents SDK

Building AI agents with OpenAI Agents SDK

1
Comments
5 min read
Evaluating RAG Systems: Measuring Retrieval Quality, Grounding, and Hallucinations

Evaluating RAG Systems: Measuring Retrieval Quality, Grounding, and Hallucinations

Comments
3 min read
Three of my agent's API calls were Opus. My logs said "200 OK" eight times.

Three of my agent's API calls were Opus. My logs said "200 OK" eight times.

Comments
2 min read
Production-Grade RAG: Why Vector Search Isn't Enough (and How Hybrid Search Fills the Gaps)

Production-Grade RAG: Why Vector Search Isn't Enough (and How Hybrid Search Fills the Gaps)

3
Comments 2
6 min read
Por que n=50.000 mentiu para mim: a armadilha estatística por trás de uma falsa vantagem setorial

Por que n=50.000 mentiu para mim: a armadilha estatística por trás de uma falsa vantagem setorial

1
Comments
5 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.