> ## Documentation Index
> Fetch the complete documentation index at: https://arizeai-433a7140.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.
# 09.29.2026: Editable Dataset Examples, Trace Metrics, and Custom Skills
> Edit dataset examples in the table and save the batch as one version, read latency, tokens, and cost on every trace tree row, and mount your own skills into MCP and PXI.
Edit a dataset's examples directly in the table and save them as one new version. Trace tree rows
now show latency, tokens, and cost, and Phoenix serves your own Agent Skills over MCP and in PXI.
# Breaking Changes in the GraphQL API
September 16, 2026
**Breaking change in arize-phoenix 20.13.0**
* **`rootSpansOnly` and `orphanSpanAsRootSpan` are removed** from `Project.spans` and
`Trace.spans`. Use `filterCondition: "parent_id is None"` instead, or `"parent_span is None"` to
count orphans as roots.
* **`Project.traceAnnotationsNames` is now `Project.traceAnnotationNames`.**
* **`applyDatasetExampleChanges` is now `patchDatasetExamples`.** It takes an ordered, JSON
Patch-style list of operations that commits as one dataset version, and requires `datasetId`.
# Edit Dataset Examples in Place
September 16, 2026
**Available in arize-phoenix 20.13.0+**
Click **Edit** above the examples table to fix input, output, and metadata JSON without opening
examples one at a time. The whole batch saves as one new dataset version.
* **Edit JSON cells in place**, add rows with an optional custom ID, and remove or restore rows
before you save. Phoenix flags duplicate custom IDs before the save goes through.
* **Save with `⌘S`.** The toolbar counts pending changes, the save dialog takes an optional version
description, and Phoenix warns before you leave the page with unsaved edits.
* **Search and split filters stay live while you edit**, and the toolbar reports any changes the
current search hides.
Ways to change an existing dataset
Group examples into named subsets
# Latency, Tokens, and Cost on Every Trace Tree Row
September 22 to 23, 2026
**Available in arize-phoenix 20.16.0+**
Each trace tree row now ends with latency, tokens, and cost, so you can spot the slow or expensive
span without opening it.
* **Hover a row to preview its breakdown.** One preview follows the pointer down the tree and loads
details only once the pointer settles, so large traces stay responsive.
* **Tokens and cost share one breakdown.** Each measure splits into token types, so you can compare
a type's share of tokens with its share of cost.
# PXI: Custom Skills, Schema Search, and Slash-Command Keys
September 21 to 29, 2026
**Available in arize-phoenix 20.15.0+ (server) and @arizeai/phoenix-cli 1.18.6+**
Point Phoenix at your own [Agent Skills](https://agentskills.io/specification) and it serves them
next to the built-in ones, both in the MCP server's instructions and in PXI's skill picker.
```bash theme={"theme":{"light":"github-light-default","dark":"github-dark-default"}}
PHOENIX_SKILLS_PATHS=/opt/skills/team-analysis,/srv/agents/skills
```
* **Each path is a skill or a directory of skills.** A skill's directory name must match the `name`
in its `SKILL.md` frontmatter, and the startup banner lists what mounted. Use absolute paths in
deployments.
* **PXI searches the GraphQL schema** instead of reading the full SDL:
`phoenix-gql schema --search ` returns ranked field signatures, and
`phoenix-gql schema --names ` prints one definition and the paths that reach it.
* **Slash-command hints respond to the keyboard** in the `pxi` terminal client. Arrow keys move the
highlight, Tab completes, and Enter runs, so `/he` plus Enter runs `/help`.
Mount your own skills and configure the endpoint
Investigate Phoenix data with PXI
# Score Classifications with Decision-Only Models
September 21, 2026
**Available in @arizeai/phoenix-evals 2.6.0+ (TypeScript)**
Pass an AI SDK evaluation model, such as TypeSafe's Jev, to any classification evaluator. Evaluation
models answer one typed choice question instead of generating text, so a single-label eval runs
faster and costs less than on a chat model.
```typescript theme={"theme":{"light":"github-light-default","dark":"github-dark-default"}}
import { typeSafeAi } from "@ai-sdk/typesafe-ai";
import { createHallucinationEvaluator } from "@arizeai/phoenix-evals";
const evaluator = createHallucinationEvaluator({
model: typeSafeAi.evaluationModel("jev-latest"),
});
const result = await evaluator.evaluate({
input: "Retrieved document: Arize Phoenix is open source.\n\nUser: Is Arize Phoenix open source?",
output: "Yes, Arize Phoenix is open source.",
});
console.log(result.label, result.metadata?.probabilities);
```
* **Same evaluators, same labels.** Phoenix detects the evaluation model and sends the call through
`experimental_evaluate` instead of `generateObject`.
* **Results carry a `label` and `score` but no `explanation`.** `metadata` reports the full label
distribution as `probabilities`, along with the `modelId`.
Trace and evaluate with TypeSafe models
Every evaluator that ships with Phoenix
# GPT-6 Sol, GPT-6 Luna, and Claude Opus 5.5
September 23, 2026
**Available in arize-phoenix 20.16.0+**
All three models are available in the playground and priced in the built-in cost manifest,
including the 272k long-context tier for the GPT-6 pair.
* **GPT-6 Sol and GPT-6 Luna are reasoning models**, so Phoenix sends them to the Responses API on
OpenAI and Azure OpenAI, where tool calls work.
* **Claude Opus 5.5 always uses adaptive thinking**, so the playground omits temperature, top-p,
top-k, and extended-thinking settings. Bedrock lists it as `anthropic.claude-opus-5-5`.
* **Opus 5.5 spans bill at Opus 5.5 rates.** They previously matched the `claude-opus-5` pattern and
billed at Opus 5's higher rates.
# TypeScript Client: Secrets, Splits, Prompt Tags, and Filters
September 21 to 29, 2026
**Available in @arizeai/phoenix-client 7.12.0+**
New helpers manage secrets, dataset splits, and prompt version tags without a hand-written REST
call, and trace and session listing now accept filter expressions.
```typescript theme={"theme":{"light":"github-light-default","dark":"github-dark-default"}}
import { createDatasetSplit } from "@arizeai/phoenix-client/datasets";
import { upsertPromptVersionTag } from "@arizeai/phoenix-client/prompts";
import { upsertOrDeleteSecrets } from "@arizeai/phoenix-client/secrets";
import { getTraces } from "@arizeai/phoenix-client/traces";
const apiKey = process.env.OPENAI_API_KEY;
if (!apiKey) throw new Error("OPENAI_API_KEY is required");
await upsertOrDeleteSecrets({
secrets: [
{ key: "OPENAI_API_KEY", value: apiKey },
{ key: "OLD_PROVIDER_API_KEY", value: null },
],
});
await createDatasetSplit({
dataset: { datasetName: "support-tickets" },
name: "regression",
exampleIds: ["ticket-14", "ticket-22"],
});
await upsertPromptVersionTag({
prompt: { versionId: "UHJvbXB0VmVyc2lvbjox" },
name: "production",
});
const { traces } = await getTraces({
project: { projectName: "my-project" },
filter: "error_count > 0 and total_cost > 0.05",
});
```
* **`upsertOrDeleteSecrets`** (7.13.0) applies a batch atomically: a value creates or updates a
key, and `null` deletes it. The response lists key names only, never values.
* **`createDatasetSplit`, `updateDatasetSplit`, and `deleteDatasetSplit`** (7.14.0) manage splits
on a dataset selected by name or GlobalID. Requires Phoenix server 19.20.0+.
* **`upsertPromptVersionTag` and `deletePromptVersionTag`** (7.15.0) create, move, and remove
prompt version tags.
* **`filter` on `getTraces` and `listSessions`** (7.12.0) takes the same expressions as the UI and
deprecates the `error`, `minLatencyMs`, and `maxLatencyMs` parameters. Requires Phoenix server
20.12.0+.
Filter vocabulary for spans, traces, and sessions
Promote a prompt version with a tag
# REST API Updates
September 23, 2026
**Available in arize-phoenix 20.16.0+**
* **`GET /v1/projects/{project_identifier}/spans` accepts `sort=start_time`**, so a page limit
returns the latest spans rather than the most recently ingested ones. The default `sort=id` keeps
insertion order.
# Additional Improvements
September 15 to 29, 2026
**Available in arize-phoenix 20.13.0+, arize-phoenix-evals 3.9.0+, and arize-phoenix-otel 0.17.2+**
* **Experiment names and descriptions are editable** from the experiments table and when you launch
an experiment from the playground.
* **`register()` no longer crashes** with `opentelemetry-exporter-otlp-proto-http` 1.45.
* **Analytics SQL on SQLite supports the bundled time functions** (`time_trunc`, `time_parse`,
`time_fmt_datetime`, and `time_sub`), and `latency_ms` is nanosecond-exact.
* **PXI streams faster**, keeps partial output when you interrupt a turn, and closes the turn's
trace when a server tool raises.
* **`PromptTemplate` message lists accept role aliases** such as `developer`, `human`, `ai`, and
`model`.
* **The rate limiter rejects an invalid initial request rate** up front.
* **SQLite autoincrement counters survive migrations** that rebuild a table, so upgrades never reuse
a primary key.
* **Metric chart time ranges stay stable**, and the table's load-more control stays in the footer in
Safari.
* **The cost manifest carries refreshed token prices** for built-in models.