LarAgent

Really excited about the release of the Laravel AI SDK - and about Taylor crafting it himself. It’s great to see interacting with LLMs and building AI features officially promoted to first-party citizenship in Laravel. Laravel is truly an incredibly productive stack for building AI-assisted software. Being a batteries-included framework, it gives you everything you need out of the box: queues and jobs, background workers, schedulers, events, caching, retries, auth, policies, pipelines, observability, and solid operational primitives. Now that Laravel officially offers an AI SDK, it’s going to be much easier for more people in the community to build AI agents, assistants, and in general AI-powered software with Laravel. This is also great news for LarAgent. Our vision with LarAgent has always been to build a full-fledged framework for agentic AI systems. But since there was nothing like this when we started, we had to implement and maintain a lot of the low-level pieces ourselves. With the Laravel AI SDK in place, we can now rely on it for core AI primitives - and focus LarAgent on higher-level concerns that matter a lot for production-grade agentic systems, especially in regulated industries like healthcare and finance: workflow orchestration, dynamic context management, evaluators, guardrails, agent interactivity, trace debugging and monitoring dashboards, and more. LarAgent 2.0 will be fully based on the Laravel AI SDK, and we’re excited about what this unlocks. We’ll also be contributing SDK-level features back to the Laravel AI SDK and introducing additional auxiliary tooling to make Laravel an even better ecosystem for building AI systems. Big kudos to Taylor and the Laravel team. This is an important step forward for the ecosystem.

We already hit v1.2.0 🥳 Is your provider failing sometimes? Automatic failover built-in 🚀 We summed up this and other great changes made after v1 was released. Article link in comments 👇 #ai #opensource #release #laragent #laravel

Multi-provider fallback for Agents. 🚀 💡 What is it? You can now define providers as an ordered array with per-provider overrides. On failures, Agents automatically switch to the next provider, including streamed responses. 🔧 Why is it important? This brings higher reliability and graceful recovery during provider hiccups. It gives teams flexibility to optimize models, keys, and costs. Backward compatibility is preserved, and deprecations are clearly signposted. ✨ The Benefits - Seamless failover across multiple providers - Per-provider config overrides with clear precedence - Backward compatible and simple to adopt 🚀 Merged and coming in the next release. Explore the updated README and prep configs for resilient Agents. ⚡ --- Contributed by: MaestroError Post generated by LarAgent itself 😄 #laragent #laravel #ai #aidevelopment #opensource

🚀 LarAgent 1.0.0 is here! Build smarter AI agents in Laravel with new reliability and tools. Perfect for production! 🎉 https://lnkd.in/eqXSbrGi

Structured Output for Claude already in LarAgent 🚀 💡 What is it? Claude in LarAgent now supports JSON Schema via output_config. Agents can request validated, strongly-typed responses, no custom parsing. 🔧 Why is it important? You get reliable schemas across streaming and tools, with clear errors. Backed by unit tests and integration checks. ✨ The Benefits - Seamless JSON Schema validation across Claude models - Full DataModel support for response structure - Reliable JSON streaming via text deltas 🚀 Rolling out in the next release. Upgrade and simplify your data extraction today. Follow for more LarAgent updates. ⚡ #Laravel #AI #Anthropic --- Contributed by: MaestroError Post generated by LarAgent itself 😄 #laragent #laravel #ai #aidevelopment #opensource

🤖 AI agents are everywhere right now, but turning an idea into something that actually works is a different story. 📹 We just dropped a new video where our CEO, Gaga Darsalia, and AI Lead, Revaz Ghambarashvili, answer the most common questions about building AI agents and our 5-week sprint process We talk through how things work in practice, from messy data and integrations to guardrails and real expectations If you’re wondering whether an AI agent could realistically help your team, this is a good place to start 👇 https://lnkd.in/d2zfm69x

We just reached 600 stars on GitHub 🥰 Thanks to all contributors and the community for your support 🙏 We are looking forward to seeing what you build with LarAgent v1.0 🚀 #opensource #laravel #laragent #aidevelopment #aiagents

🚀 Now that LarAgent 1.0.0 is officially released, it felt like the right time to go a bit deeper. 👀 This article breaks down the key changes in v1.0: from reliability and context management to developer experience and enterprise readiness. If you’re using LarAgent or considering it for your next Laravel app, this should give you a clearer picture of what’s new and why it matters 👇 🔗 https://lnkd.in/dnjazBiX

This major release takes LarAgent to the next level 🚀 focused on: - structured responses - reliable context management - production-grade agent behavior Designed for teams and applications where predictability, observability, and scalability matters 💪 Check article in comments 👇 #laravel #laragent #aiagents #update #majorrelease #opensource