> ## 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.
# 08.04.2026: Chat Completions Proxy, AI Query, and Annotation Charts
> An OpenAI-compatible chat completions endpoint backed by server-held credentials, plain-English filter composition, per-annotation metric charts with no selection cap, a conversation-grounded Hallucination evaluator, annotations in span downloads, and a pinned note bar in span details.
# OpenAI-Compatible Chat Completions Endpoint
August 4, 2026
**Available in arize-phoenix 19.16.0+**
Phoenix now exposes `POST /v1/chat/completions` in the OpenAI wire format. Point any OpenAI-compatible
client at your Phoenix server and Phoenix proxies the call to the provider you name, resolving the
provider credentials on the server — the secret store first, the process environment second. Callers
authenticate to Phoenix and never handle provider API keys.
* **Model IDs name the provider** — `{provider}:{model_name}` for a built-in provider
(`openai:gpt-4o`, `anthropic:claude-sonnet-4-5`), or `custom:{provider_id}:{model_name}` for a
[custom provider](/docs/phoenix/settings/custom-ai-providers) record stored in Phoenix. Only the
first colons are split, so model names that contain colons survive intact.
* **Streaming** — set `stream: true` for server-sent `chat.completion.chunk` events terminated by
`data: [DONE]`. `stream_options: {include_usage: true}` appends a final usage chunk.
* **Familiar parameters** — `temperature`, `top_p`, `max_tokens` / `max_completion_tokens`, `stop`,
`seed`, `frequency_penalty`, and `presence_penalty` all pass through.
* **OpenAI-shaped errors** — every failure, including validation errors, comes back as
`{"error": {"message", "type", "code"}}`. Provider HTTP errors forward their status; an
unreachable provider becomes a `502`.
* **Available to every authenticated role** — the endpoint writes nothing, so viewers can call it.
```bash theme={null}
curl -X POST "$PHOENIX_HOST/v1/chat/completions" \
-H "Authorization: Bearer $PHOENIX_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai:gpt-4o",
"messages": [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Say hello."}
]
}'
```
Because the wire format is OpenAI's, the OpenAI SDKs work unchanged:
```python theme={null}
import os
from openai import OpenAI
client = OpenAI(
base_url=f"{os.environ['PHOENIX_HOST']}/v1",
api_key=os.environ["PHOENIX_API_KEY"],
)
completion = client.chat.completions.create(
model="anthropic:claude-sonnet-4-5",
messages=[{"role": "user", "content": "Summarize what a span is in one sentence."}],
)
print(completion.choices[0].message.content)
```
Tool calling, `n > 1`, and non-text `response_format` are rejected with a `400`.
Store provider connection details and credentials on the server
Manage the credentials the proxy resolves
# AI Query for Filter Fields
August 4, 2026
**Available in arize-phoenix 19.17.0+**
Describe what you want in plain English and let Phoenix write the filter expression. The filter field
above the spans and traces tables — and above the experiment runs table — gains a sparkle toggle that
switches it into plain-English mode.
* **Enter converts** — in plain-English mode, Enter translates your prose into a filter expression,
streaming it in as it forms. From expression mode, `⌘`/`Ctrl`+Enter hands the current draft to AI
query directly.
* **Validated, with one correction round** — the generated expression goes through the same validator
the field itself uses, and the model gets one chance to fix a rejected expression before you see it.
* **Escape undoes** — Escape walks back whatever AI query last did, restoring your original phrasing.
* **The model's vocabulary is the field's vocabulary** — the field names and examples handed to the
model are derived from the same completions and snippets that power the typeahead, so the two can't
drift apart.
Pick the model on the new **Profile → Generative AI** page, or from the gear popover on the filter
field itself:
* **Browser AI** — the browser's built-in on-device model (Chrome and Edge's Prompt API). No
credentials, no network round trip, and the default wherever a built-in model is available. A
companion card shows download status and can fetch the model ahead of first use.
* **Any provider Phoenix knows** — built-in providers, Azure, Bedrock, and stored custom providers,
all called through the new `/v1/chat/completions` proxy so no API key ever reaches the browser.
Only your query and the filter field's vocabulary are sent to the model.
The span filter expression language AI query writes
# Annotation Metric Charts and Uncapped Chart Selection
July 30 – August 3, 2026
**Available in arize-phoenix 19.11.0+ (project charts), 19.13.0+ (experiments), 19.15.0+ (deferred annotation charts)**
Evaluation results now get first-class charts, and the three-chart limit on the chart strip is gone.
* **A chart per annotation name** — the project **Metrics** page adds span, trace, and session
annotation sections, each followed by a grid with one chart per annotation name on that level.
Every chart plots mean score over time, with a score/label toggle for annotations that carry both.
* **Per-annotation charts in the chart strip** — the same charts are selectable in the **Charts** menu
above the spans, traces, and sessions tables, alongside the overall annotation charts.
* **No selection cap** — the chart strip above the project tables and above the experiments table no
longer limits you to three charts. Pick as many as you want to read.
* **Charts load when you reach them** — chart panels render a skeleton until they scroll into view,
and a chart whose panel is hidden freezes its query inputs instead of refetching. A page full of
annotation charts no longer fires every query at once.
The per-project metrics dashboard
Produce the annotations these charts summarize
# Conversation-Grounded Hallucination Evaluator
August 3, 2026
**Available in arize-phoenix 19.14.0+ (built-in evaluator) and @arizeai/phoenix-evals 2.2.0+ (TypeScript)**
The Hallucination evaluator now judges an assistant response against **the conversation it came from**
— earlier turns, tool calls, and the results those tools returned — rather than against a separately
supplied context block. Use it for multi-turn agents where the source of truth is the transcript;
reach for Faithfulness when you have one retrieved context block.
* **`input` and `output` only** — `input` is the full record the assistant had available (its last
message is the turn being answered) and `output` is the response being judged. There is no longer a
separate `context` field.
* **Catches fabricated work** — invented specifics, tool results a tool never returned, findings from
material that was never read, and actions reported as already done.
* **Absence of evidence counts** — a confident, fluent response that asserts situation-specific facts
absent from the input is hallucinated. Ordinary general knowledge is exempt.
* **As a built-in Phoenix evaluator** it is promoted in the dataset evaluator gallery, and the
`output` it judges includes the span's tool calls.
```typescript theme={null}
import { createHallucinationEvaluator } from "@arizeai/phoenix-evals";
import { openai } from "@ai-sdk/openai";
const hallucinationEvaluator = createHallucinationEvaluator({
model: openai("gpt-4o"),
});
const result = await hallucinationEvaluator.evaluate({
input: [
"User: What's our refund window?",
"Tool (lookup_policy): Refunds: 30 days from delivery.",
"Assistant: 30 days from delivery.",
"User: And for electronics?",
].join("\n"),
output: "Electronics can be returned within 90 days.",
});
console.log(result.label); // "hallucinated"
```
**Labels changed.** The evaluator now returns `grounded` / `hallucinated` instead of
`factual` / `hallucinated`, and scores are minimized (`hallucinated` = 1, `grounded` = 0). Stored
evaluations, dashboards, thresholds, and label filters built on the previous evaluator may need
migrating and should not be compared directly with new results.
Full input formatting guidance and usage examples
Ground a response in a single retrieved context
# Annotations in Span Downloads
August 1, 2026
**Available in arize-phoenix 19.13.0+ (annotations) and 19.16.0+ (streamed downloads)**
Exported spans can now carry their evaluations with them. The **Download selection** dialog adds
**Include span annotations** and **Include trace annotations** checkboxes, both on by default.
* **OpenInference semantic attributes** — span annotations are attached to their own span and trace
annotations once per trace (on the root span where there is one), as indexed `annotations.*` and
`trace.annotations.*` attributes carrying name, annotator kind, score, label, explanation,
identifier, and JSON-encoded metadata.
* **Streamed straight to disk** — where the browser supports it, a download writes through a save
file picker as pages arrive instead of buffering the whole export in memory.
* **Parallel fetches** — independent ID batches are fetched with bounded concurrency while cursor
pagination within a batch stays ordered, so large selections finish substantially faster.
# Pinned Note Bar in Span Details
July 30, 2026
**Available in arize-phoenix 19.11.0+**
Reviewing a trace and want to write down what you found? Press `n` in span details — or use the
toggle on the **Notes** card — and a note bar rises from the bottom of the pane and stays there as
you move between spans.
* **Enter adds the note**, Shift+Enter starts a new line, and the field grows to six lines before it
scrolls.
* **Escape closes** an empty bar; with a draft in it, Escape just blurs so nothing is lost.
* **Stays open across spans** — the bar is remembered as a preference, so a review session keeps its
note field until you close it. A failed submission puts your draft back.
# Also in This Release
July 30 – August 4, 2026
**Available in arize-phoenix 19.11.0–19.17.0 and @arizeai/phoenix-cli 1.13.1–1.14.0**
* **The time range follows you into a project** — opening a project from the projects list carries the
list's time range along, instead of resetting to the default window (arize-phoenix 19.16.0+).
* **Collapsed cards say what they hold** — LLM messages, invocation parameters, LLM input, and
playground and prompt chat templates show a one-line excerpt of their body in the header while
collapsed (arize-phoenix 19.16.0+).
* **Exception stack traces are readable** — an `exception` span event renders its
`exception.stacktrace` as a dedicated, expandable **Stack trace** card with a copy button, with the
remaining attributes below it (arize-phoenix 19.16.0+).
* **Tool counts in LLM span card headers** — the input card subtitle shows how many tools the model
had available, and the output card shows how many tool calls it made. Every span card also gets a
copy button in its top-right corner (arize-phoenix 19.11.0+).
* **Reasoning survives OTel conversion** — reasoning parts in OTel GenAI `gen_ai` messages are now
flattened to OpenInference message contents with type `reasoning` instead of being dropped
(arize-phoenix 19.14.0+).
* **Stricter span filter validation** — malformed filter conditions, unsupported syntax, and
excessively nested expressions are rejected with a clear syntax error rather than failing
unpredictably (arize-phoenix 19.11.1+).
* **Provider-agnostic model cost entries** — creating a model in **Settings → Models** accepts an
empty provider, and a name collision now reports the actual conflict (arize-phoenix 19.11.0+).
* **Refreshed built-in token prices** so cost tracking stays accurate for the current model lineup
(arize-phoenix 19.11.1+).
* **`px setup` fails when verification fails** — a run that tried to confirm traces and never saw one
exits `6` (`NOT_VERIFIED`) instead of `0`, so `px setup && npm run dev` and `set -e` bootstrap
scripts stop on a broken instrumentation. Choosing to verify later still exits `0`.
* **`px` suggests upgrading** — an unknown command now compares your installed version against the
latest published one and points at `px self update` when you're behind, or `px --help` when you're
current (@arizeai/phoenix-cli 1.14.0+).
* **`px auth status`** no longer errors when a profile holds stale OAuth credentials but the server
has since allowed anonymous access (@arizeai/phoenix-cli 1.13.1+).