arize-phoenix

@ArizePhoenix

Open-Source AI Observability and Evaluation

Notebook or Container

github.com/Arize-ai/phoen…

Joined February 2023

319 Following

1,794 Followers

Replies Reposts Media

arize-phoenix
@ArizePhoenix
Oct 16
Why make an LLM read thousands of spans to count errors? In our benchmark, querying traces with SQL + code mode cost ~17x less than fetching spans page by page. See how Arize Phoenix MCP does it, live in Claude Code and Cursor. Register here:
Inside Phoenix MCP: how remote MCP, SQL, and code mode let agents query traces · Luma
From luma.com
1
arize-phoenix
@ArizePhoenix
Oct 15
Changing one tool description reduced Codex’s unnecessary skill loads from 51 to 0 and lowered the cost of our benchmark run from $11.85 to $8.34 The walkthrough shows how to benchmark AI agents and tools with Harbor and Arize Phoenix
Arize AI
@arizeai
Oct 12
We benchmarked 5 agent configurations across 18 tasks and 270 sandboxed runs with @ArizePhoenix. Three configurations answered every task correctly. But Codex loaded an unnecessary skill in 51 of 54 attempts. The benchmark score said everything was fine, but our agent traces
1
arize-phoenix
@ArizePhoenix
Oct 14
You can now trace Claude Managed Agents in Arize Phoenix. Learn more: claude.com/resources/arti…
ClaudeDevs
@ClaudeDevs
Oct 14
Now in public beta, Claude Platform can send OpenTelemetry traces from the Messages API, Claude Managed Agents, and the Claude SDKs to your observability stack.
00:00
1
arize-phoenix
@ArizePhoenix
Oct 10
New compare view for traces on projects Analyze trajectory differences between two agent turns side by side.
arize-phoenix
@ArizePhoenix
Oct 9
new baselining capabilities for experiments Now when you have a baseline set on your experiments or when you care comparing two experiments, you can see the deltas of your evals, tokens, latency and cost so you can quickly identify regressions and improvements.