AI Queries and Browser AI Phoenix filters now understand English. Type "responses with apologies" and get a real filter expression back: ▎ 'sorry' in output.value or 'apolog' in output.value

The experiment said the fix works, but experiments aren't production. Real spans, new prompt, zero errors. Will PXI get it right? youtu.be/biMD5LHdXQk

Balancing cost, speed, and correctness can be a tricky balance. That's why we need good visuals to figure out the right sweet spot!

Customizable visualizations of your agent traces are here. Track cache hits, online eval degradations, tool call errors, all in real-time so you can quickly identify problems in production. What does your agent operations center look like?

New Sandbox in Phoenix - Monty (Python)!

In an agent session, a cache miss re-bills your entire history at full input price. That's why a "continue" after a coffee break can cost more than the model's actual answer. Cache read vs. write isn't a footnote in your bill. Earendril's post is a must read. earendil.com/posts/promp...

The agent kept writing SQL querying a "products" table but the table is called "catalog." PXI analyzes the problem and proposes a fix all on its own: youtu.be/CN8VwB_4V_Y

Search -> Select -> Download -> Coding Agent. Feed critical traces back to Agents. No fancy words for this one: this is plain old-fashioned debugging. Just now with Agents.

Your agent's SQL tool calls start failing with "no such table: products." Before you attempt a fix, take a snapshot of the failures with a dataset. Just ask PXI to do it for you: youtu.be/gb_78l788Ls

Making a Dataset from Failing Traces with Phoenix and PXI

An agent's SQL tool calls start erroring out with "no such table: p...

youtube.com

phoenix is now is an oauth2 authorization server! `px auth login` opens your browser, you get a short-lived user-scoped token. no long-lived api keys. and admins can audit / revoke every cli and mcp grant from one screen. remote mcp server ships in beta too - more on this soon.

Phoenix's agent PXI can propose annotation categories, annotate traces en mass, and even suggest fixes based on patterns of failures. Learn how to do it: youtu.be/iF25CqJv4tA

Our favorite tools are the ones that have maximum customizability. Last week we added customizable charts, command K, and recent searches. this week we've added custom column ordering. Built to help you have the tables and dashboards you need to monitor agents day in and day out.

What does your agent operation center look like?

Experiment Baselining and Charts When trying to determine if a new model is up to the task, you need to factor in many dimensions. performance - measured by evals latency - is the model fast enough to give you the right UX tokens - how chatty is the model to achieve the result

Do you love localfirst development? @nearestnabors.com will be speaking on how to get frontier LLM results on device using Phoenix and prompt engineering techniques on Sunday, July 12, 14:15. See you soon, Berlin! www.localfirstconf.com/

Local-First Conf 2026

Join us for the third edition of Local-First Conf. Connect with a rapidly-growing community in an intimate setting. Berlin 12-14th July 2026.

localfirstconf.com

Phoenix has an agent built into it now! PXI can help you find the crucial traces you should actually be reading. Short video on how to do use PXI in your daily flow: youtu.be/5lUgdRFf4ZI

Find the traces that matter with Phoenix and PXI

Hundreds of traces and they all look fine? 🤔 Here's how to find the...

youtube.com

⌘K is in Phoenix. jump to any project, dataset, prompt, or experiment. No mouse required.

Agent traces and trajectories are growing increasingly longer and more complex. We've seen some traces 1000s of spans deep. That's why we've added trace search. Search across a trace and the UI will now show you the call stack to the spans you are looking for across workflows and sub-agent calls.

The OSS team has a mantra: 3 clicks to glory. We're not there yet but we're getting closer!

Fable 5 might be amazing, but it can't tend the grill 🍔 Happy Fourth of July from our team to yours. We hope you are getting to spend the day with friends, family, and loved ones.

PXI now supports sub-agent streaming so you can inspect the execution of the sub-agents you kick off. We've started moving much the skills and tools available in the main agent into the sub-agents to empower interesting delegation patterns.

PXI (Phoenix Intelligence) now runs in your terminal! You can now use PXI without leaving your terminal. It's the same agent that powers the in-browser experience, now available as an interactive chat in your shell. npm install -g @arizeai/phoenix-cli@latest > pxi

📦🏷️🔖🔑 This week in Phoenix: - server-side bash for PXI subagents, - label management on list pages, - trace-level annotations everywhere - and OAuth2 role-override preservation. arize.com/docs/phoeni...

Meet PXI (pronounced "pixie") 🎉 the AI engineering agent we built into Phoenix. Hand it the investigation instead of scrolling through traces by hand, and it works through your telemetry the way a coding agent works through code. The full story of how we built it 👉 arize.com/blog/meet-pxi/

Meet PXI: the AI engineering agent inside Phoenix

PXI is the open-source AI engineering agent built into Phoenix. Hand it a failing trace, an evaluator, or a prompt, and it investigates your telemetry for you.

arize.com

📊 Phoenix 17.7.0 makes your token usage legible. New token detail charts break prompt + completion tokens into their parts, over time: • Prompt → input, cache read, cache write, audio • Completion → output, reasoning, audio

Phoenix just hit 10,000 GitHub stars! Three years ago, Phoenix didn't exist. Arize was a closed-source company. A small team was asked to change that. Catch the full interview with the team who made it happen and where AI observability is going next: arize.com/phoenix-10k

"Don't trust. Evaluate." @nearestnabors.com set out to replace Sonnet with Gemma. The evals showed a quantifiably better option. Full walkthrough: capability evals + prompt engineering to ship a local 3B that matches Sonnet, 2x faster, $0/call. Built with Phoenix. arize.com/blog/how-to...

Phoenix now lets you compose evaluation strategies in code. Most eval tooling hands you a fixed menu of judge templates. Real evaluation is rarely that tidy.

The Arize DevRel team wants to connect with Phoenix users like you. What you're tracing, what's working, what's rough? Schedule time with the team here: cal.com/team/arize-...

Phoenix and AX Free Weekly User Interviews | Arize Dev Rel | Cal.com

Phoenix and AX Free Weekly User Interviews

cal.com

Something we’ve been playing with and liking a lot: Give every coding agent its own observability stack.