DEV Community

# llm

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Qwen3.6 MoE, WritHer Offline AI, & llama.cpp Benchmarks Lead Local AI News

Qwen3.6 MoE, WritHer Offline AI, & llama.cpp Benchmarks Lead Local AI News

Comments
3 min read
How to Detect If Your LLM Proxy Is Silently Eating Your Tokens

How to Detect If Your LLM Proxy Is Silently Eating Your Tokens

Comments
5 min read
Subliminal Learning and the Hidden Channel Problem in LLM Training

Subliminal Learning and the Hidden Channel Problem in LLM Training

Comments
2 min read
Why Your 5-Agent System Forgets State (And How to Fix It)

Why Your 5-Agent System Forgets State (And How to Fix It)

Comments
1 min read
qwen3.6 scores 73.4 on SWE-bench with only 3B active parameters. here's why that matters.

qwen3.6 scores 73.4 on SWE-bench with only 3B active parameters. here's why that matters.

Comments
4 min read
10 Ways To Reduce Your LLM API Costs

10 Ways To Reduce Your LLM API Costs

8
Comments 1
7 min read
How Agentic Search Actually Works: The Research Loop Link-Fetching Agents Miss

How Agentic Search Actually Works: The Research Loop Link-Fetching Agents Miss

Comments
4 min read
Gemma 4 wrote three summaries in one response. The middle one was a self-disclaimer.

Gemma 4 wrote three summaries in one response. The middle one was a self-disclaimer.

2
Comments 1
6 min read
Prompt Hashing for Duplicate Detection: Cutting LLM Waste With SHA-256

Prompt Hashing for Duplicate Detection: Cutting LLM Waste With SHA-256

Comments
4 min read
Bringing Generative AI to Microcontrollers: Introducing NocLLM

Bringing Generative AI to Microcontrollers: Introducing NocLLM

1
Comments
3 min read
AI Red-Teaming for Beginners: Where to Start and What to Test

AI Red-Teaming for Beginners: Where to Start and What to Test

Comments
5 min read
Runware: One API for All AI Modalities — AI University Update (77 Providers)

Runware: One API for All AI Modalities — AI University Update (77 Providers)

1
Comments 1
2 min read
How to Compare AI Models Without Getting Fooled by Benchmarks

How to Compare AI Models Without Getting Fooled by Benchmarks

10
Comments
2 min read
AI 週報 — 2026/04/10–2026/04/17 模型封鎖潮來了,但工具鏈才是真戰場

AI 週報 — 2026/04/10–2026/04/17 模型封鎖潮來了,但工具鏈才是真戰場

Comments
1 min read
Google I/O Review (1/5) — Gemini 3.5 'Flash' Costs 15x More Than Flash 2.0. It's Pro in Disguise

Google I/O Review (1/5) — Gemini 3.5 'Flash' Costs 15x More Than Flash 2.0. It's Pro in Disguise

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