Arize Phoenix rolled out several capabilities in its latest release.
Phoenix's new support for LLM traces and spans means that AI engineers and developers can get visibility at a span-level and see exactly where an app breaks, with tools to analyze each step rather than just the end-result.
This capability is particularly useful for early app developers because it doesn't require them to send data to a SaaS platform to perform LLM evaluation and troubleshooting -- instead, the open-source solution provides a mechanism for pre-deployment LLM observability directly from their local machine. Phoenix supports all common spans and has a native integration into LlamaIndex and LangChain.
The new Phoenix LLM evals library is also designed for fast and accurate LLM-assisted evaluations, ultimately making the use of the evaluation LLM easy to implement. Applying data science rigor to the testing of model and template combinations, Phoenix offers proven LLM evals for common use cases and needs around retrieval (RAG) relevance, reducing hallucinations, question-and-answer on retrieved data, toxicity, code generation, summarization, and classification. The Phoenix LLM evals library is optimized to run evaluations quickly with support for the notebook, Python pipeline, and app frameworks such as LangChain and LlamaIndex.
"Large language models are poised to transform industries and society, but when it comes to robust performance going from toy to production remains a challenge," said Jason Lopatecki, CEO and Co-Founder of Arize AI. "These industry-first updates from Phoenix promise to provide better LLM evals and deeper troubleshooting to make complex LLM-powered systems ready and reliable in the real world."
The Latest
For years, DevOps teams operated under a simple assumption: collect enough telemetry, and you can find and fix any problem. That assumption is breaking down. Modern enterprises now operate across microservices, hybrid cloud environments, APIs, Kubernetes, and highly automated delivery pipelines. Releases happen continuously, dependencies shift constantly, and failures spread faster than teams can diagnose them ...
New Relic surveyed IT and engineering leaders from the media and entertainment (M&E) sector to understand what's working — and where challenges persist with their observability practices. The findings reveal how M&E organizations are navigating rising platform complexity, audience expectations, and AI-driven change. Below are five takeaways that stand out ...
Let me start with something I've seen play out more times than I can count. A team hits a wall with the cloud. Costs creep up, then spike. Performance starts to feel inconsistent. Someone in finance asks a simple question like "why did this double?" and nobody has a clean answer ... Maybe this isn't the right place for everything. That realization feels like a breakthrough, like you've identified the problem. In reality, you've just identified the starting line ...
In MEAN TIME TO INSIGHT Episode 24, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses network observability tool sprawl ...
In cloud-native systems, scaling is often as simple as moving a slider. For on-premise databases, the stakes are different. Over-provisioning hardware is expensive. Under-provisioning leads to performance bottlenecks that are difficult to fix once the equipment is in the rack ...
When most people think about cybersecurity, they picture firewalls, encryption, and access controls — technical tools designed to protect systems and data. But beneath the technology lies a deeper set of principles about trust, decision-making, and resilience ... The best leaders don't eliminate risk. They manage it intelligently. And in many ways, cybersecurity offers a surprisingly useful playbook for doing exactly that ...
Many organizations assumed their infrastructure strategy was settled. It had been implemented, optimized and built into long-term plans. Recent changes in technology and vendor consolidation are forcing a second look. Cloud outages and licensing changes have exposed how much dependency exists on a small number of platforms. As a result, organizations are reevaluating whether those decisions still hold up under current conditions ...
Edge AI is strategically embedded in core IT and infrastructure spending across industries, according to the 2026 Edge AI Survey from ZEDEDA. The research shows that 83% of C-suite and IT executive respondents say edge AI is important to their core business strategy ...
As AI adoption accelerates, operational complexity — not model intelligence — is becoming the primary barrier to reliable AI at scale, according to the State of AI Engineering 2026 from Datadog ... The report highlights a compounding complexity challenge as AI systems scale ... Around 5% of AI model requests fail in production, with nearly 60% of those failures caused by capacity limits ...
For years, production operations teams have treated alert fatigue as a quality-of-life problem: something that makes on-call rotations miserable but isn't considered a direct contributor to outages. That framing doesn't capture how these systems fail, and we now have data to show why. More importantly, it's now clear alert fatigue is a symptom of a deeper issue: production systems have outgrown the current operational approaches ...