DEV Community

#llm

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Thinking On or Off? Gemma 4 E4B on an 8 GB RTX 3050

Thinking On or Off? Gemma 4 E4B on an 8 GB RTX 3050

7
Comments
9 min read
Stop Paying LLM Prices for an if Statement: A Field Guide to Decision Models in Late 2026

Speed claims face skepticism

Stop Paying LLM Prices for an if Statement: A Field Guide to Decision Models in Late 2026

12
Comments 6
16 min read
Can a plain table hijack an AI agent?

Facts bypass instruction filters effortlessly

Can a plain table hijack an AI agent?

5
Comments 13
4 min read
I built a RAG that refuses to answer — and measured when it should

I built a RAG that refuses to answer — and measured when it should

1
Comments 3
2 min read
Your Model Fits in VRAM. Your Context Window Might Not

Your Model Fits in VRAM. Your Context Window Might Not

1
Comments
3 min read
The Flag That Fixed My Latency Was Quietly Ignored for Three Weeks

The Flag That Fixed My Latency Was Quietly Ignored for Three Weeks

Comments
3 min read
I tested the idea behind Claude Code's new Code Mode. The AI read 11x fewer tokens and stopped miscounting

I tested the idea behind Claude Code's new Code Mode. The AI read 11x fewer tokens and stopped miscounting

Comments
8 min read
Do llms.txt and schema markup change which pages AI engines cite? What the citation data shows

Do llms.txt and schema markup change which pages AI engines cite? What the citation data shows

Comments 2
6 min read
Send one prompt to GPT, Claude, Gemini and Grok with 40 lines of Node (and the reasoning-token trap I hit)

Send one prompt to GPT, Claude, Gemini and Grok with 40 lines of Node (and the reasoning-token trap I hit)

Comments
2 min read
Qwen3.5-9B on RTX 5060 Ti: 260k tokens of context in 16 GB

Qwen3.5-9B on RTX 5060 Ti: 260k tokens of context in 16 GB

Comments
4 min read
I ran the same 10 prompts through GPT, Claude, Gemini and Grok. Here is the pattern.

I ran the same 10 prompts through GPT, Claude, Gemini and Grok. Here is the pattern.

Comments
3 min read
Tokens and context windows: why your AI bill is 19 times bigger than you expected

Tokens and context windows: why your AI bill is 19 times bigger than you expected

Comments
4 min read
What ISO 42001 audit prep actually takes: 100+ AI projects from the provider side

What ISO 42001 audit prep actually takes: 100+ AI projects from the provider side

Comments
4 min read
Sizing a looped LLM: SenseNova-Loop-9B stores 32 layers, runs 56, and fits a 20 GB card

Sizing a looped LLM: SenseNova-Loop-9B stores 32 layers, runs 56, and fits a 20 GB card

Comments
4 min read
Why I Run AI at Home, and LM Studio Tips for Beginners

Why I Run AI at Home, and LM Studio Tips for Beginners

Comments
4 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.