cd/sources/dev-to· home› sources› Dev.to
cat /sources/dev-to.feed | wc -l → 26846

Dev.to

articles 26846 domain dev.to → page 127/1343 feed RSS
16:26
2026-10-21
dev.to
artificial-intelligence

Stop Asking Video Models to Run Physics: Inside Matrix-Game 4.0

Skywork AI and partner institutions released Matrix-Game 4.0 (arXiv:2610.27999), a neural game world model that splits simulation from rendering by keeping world state in a lightweight traditional gam…

15:59
2026-10-21
dev.to
ai-agents

Idempotency for Agent Actions

A developer at Trida AI outlined a design for making agent tool calls idempotent, arguing that retried write operations—refunds, orders, tickets—can duplicate side effects because agents retry from mo…

15:08
2026-10-21
dev.to
ai-infrastructure

Build vs Managed Vector Database: A Total-Cost Framework

A developer published a total-cost framework comparing managed vector databases against self-hosting, finding that for workloads under roughly 200 million vectors, managed services like Weaviate Cloud…

15:03
2026-10-21
dev.to
ai-agents

Installs Claude Code Mods with one command

A catalog of 38 Claude Code "Mods" has launched on aitmpl.com, letting developers install function-hook plugins with a single npx command instead of packaging and publishing them through a plugin mark…

14:31
2026-10-21
dev.to
artificial-intelligence

GPT-6.1 Sol: A Practical Guide

OpenAI released GPT-6.1 Sol, an intermediate reasoning model positioned between GPT-6 Luna and flagship GPT-6 Astra, offering near-Astra performance on software engineering, computer use, and professi…

14:31
2026-10-21
dev.to
ai-agents

The Definitive Guide to the Claude Agent SDK

Anthropic's Claude Agent SDK, the programmatic engine behind Claude Code, packages an autonomous agent harness into a TypeScript and Python client library, handling multi-turn tool loops, automatic co…

14:28
2026-10-21
dev.to
ai-safety

AI red teaming methodology for customer-facing chatbots

ClearPath Security published a methodology for AI red teaming customer-facing chatbots, framing the exercise as a scenario-led assessment of how assistants handle hostile, misleading, or malformed inp…

14:28
2026-10-21
dev.to
ai-agents

Burn less tokens

A developer released lessloc, an open-source approach for instructing coding agents to produce less code and fewer comments, aimed at reducing token consumption and context bloat. The developer report…

← prev page 127 / 1343 next →