3.10.0 (2026-11-10) Features decision-models: support OpenAI Decisions API in Python classification evaluators (f6dc888) evals: support TypeSafe Jev decision models in Python classification evaluators (9089efc) evals: use Anthropic native structured output in AnthropicAdapter (2fffecb) Bug Fixes evals: ask for the Anthropic tool call instead of forcing...
Arize Phoenix client v3.7.0 adds model parameter and agent setting recording to experiment metadata and includes validation for annotation dataframes.
Arize Phoenix v21.0.0 adds a system state table, new annotation kinds for decisions and agents, and runtime cost tracking for models.
Arize Phoenix v20.27.0 adds support for TypeSafe Jev decision models, improves PR review grading, and includes security fixes for GitHub token handling.
Arize Phoenix v20.26.0 adds session ID linking to Claude Console, factory candidate nomination, and Pipecat tracing support.
20.24.0 (2026-10-21) Features report uncaught frontend errors to the server log (2c0873c) Bug Fixes deps: exclude compromised litellm 1.107.0 (f6a0231) deps: require certifi 2026.10.18 to distrust the compromised emSign root (9056a4a) deps: require Node releases patched in the October 2026 security release (b7d454d) Documentation add Atomic Agents to the...
20.23.0 (2026-10-16) Features server: export server logs over OTLP (add77bf) Bug Fixes playground: forward Bedrock tool results directly as text (15355fd) Documentation add Atomic Agents tracing integration (fe5e933) Add Phoenix release notes , 2026-10-14 (97c4b2a)
20.22.0 (2026-10-15) Features playground: add Meta muse-wonder-1.0 and gemini-4-argon (d5168d1) Bug Fixes oauth2: preserve SameSite=Strict cookies on server-owned return_url after IdP sign-in (c332b0c) Documentation add Claude Managed Agents tracing docs (54fe899) add standalone release note for the Harbor plugin (fe4a06b)
0.2.0 (2026-10-12) Features phoenix-sqlean: bundle SQLite 3.54.0 and support Python 3.15 (fff25bc)
20.21.0 (2026-10-11) Features assign a random gradient to projects created by ingestion (8946f7a) cost-tracking: support Mistral AI model pricing (0e33736), closes #16878 Bug Fixes agents: upgrade pydantic-ai for CVE-2026-107286 (3cdb004) deps: update arize-phoenix-evals to 3.9.2 (e8aa1ea) playground: sync built-in model prices and model lists (35f86e5)...
20.20.0 (2026-10-08) Features harbor: prompt hill-climb task on a text-to-SQL dataset (bd78550) mcp/sql: allowlist the annotation, evaluator, prompt, split and job tables (a37b841) mcp: add GraphQL schema, query, and mutation tools (9351580) playground: add claude-haiku-5-5 and remove retired Anthropic models (b978e7c) pxi: make PXI's MCP code mode...
20.19.0 (2026-10-01) Features datagen: send each recorded application to its own project (d99a4db) tracing: add the DECISION span kind across the platform (43d1a54) Bug Fixes honor external resources flag for WASM and UI (7a71a21) Documentation remove legacy evaluation cookbooks (4506880)
20.17.0 (2026-09-30) Features agents: mark bash tool spans as errors on non-zero exit codes (cd0512e) graphql: add sort and sequenceNumbers to Dataset.experiments (9c14c5a) rest: add sequence_number, sort_dir, and sequence_numbers to list_experiments (305e1a9) Bug Fixes agents: leave the user out of the internal GraphQL scope when no one is logged in...
20.16.0 (2026-09-23) Features add gpt-6-sol, gpt-6-luna, and claude-opus-5-5 to the cost tables and the playground (59d09f4) js: add dataset split write helpers (aab229f) js: add secrets management helper (c282563) spans: sort the span list endpoint by start_time (03f62e1) Bug Fixes agents: close out the PXI turn trace when a server tool raises (3a3d57b)...
3.9.0 (2026-09-21) Features evals: add retrieval relevance evaluator (7efa7b2) evals: add session-level pii_detection evaluator (0e3b2c2) evals: completeness evaluator (63805d0) evals: deprecate document relevance evaluators (d67ea3f) Bug Fixes evals: accept role aliases in PromptTemplate message lists (767847c) evals: use non-blocking sleep in async rate...
20.14.0 (2026-09-18) Features evals: Harbor benchmark for Phoenix MCP and CLI tools (7f7d382) Bug Fixes cost: update built-in model token prices (f4c3bae) pxi: link references from their owning skill (4de212c) Documentation server: align create_graphql_router's docstring with its signature (81c8e25) Miscellaneous Chores release arize-phoenix 20.14.0...
20.13.0 (2026-09-16) Features datasets: editable dataset examples table (705ef38) db: use sqlean time extension for date_trunc and latency_ms on SQLite (5269488) graphql: express patchDatasetExamples as an ordered JSON Patch-style operation list (c7aa7c0) graphql: remove rootSpansOnly in favor of the span filter DSL and document the DSL in the skills...
20.12.0 (2026-09-14) Features api: expose trace and session filter expressions (d613de7) server: add GET /datasets/{dataset_identifier}/splits (e60697f), closes #12090 Bug Fixes cost: update built-in model token prices (25b95d7) server: address review findings for GET /datasets/{dataset_identifier}/splits (655802b) server: scope listed dataset splits by...
20.11.0 (2026-09-12) Features annotations: add GraphQL mutations for span, trace, and session notes (e374a4d) mcp: serve the shared skills root on the /mcp mount (dd2e22d) skills: add phoenix-error-analysis skill (df349f2) Bug Fixes mcp: carry lifespan state into in-process tool dispatch (a516ac3) Documentation restructure the Claude Code plugin docs as its...
3.8.0 (2026-09-11) Features evals: add retrieval relevance evaluator (7efa7b2) evals: add session-level pii_detection evaluator (0e3b2c2) evals: completeness evaluator (63805d0) evals: deprecate document relevance evaluators (d67ea3f) Bug Fixes evals: use non-blocking sleep in async rate limiter (2ecfd47) upgrade to anthropic SDK v1 (b28cf94)
3.7.0 (2026-09-10) Features evals: completeness evaluator (63805d0) evals: deprecate document relevance evaluators (d67ea3f)
3.5.0 (2026-09-08) Features add MiniMax provider integration (7e01ce7) client: add ATIF tracing to Harbor plugin (30dccec) client: add error and latency filters to get_traces in python and typescript clients (c839254) Bug Fixes annotations: validate all configured ID columns, not just the first (7654185) ci: format client docs and skip fork PR comments...
3.4.0 (2026-09-03) ⚠ BREAKING CHANGES server: GET /v1/model_providers no longer returns custom providers or a next_cursor, and built-in entries expose provider rather than kind + provider_key. Custom providers move to GET /v1/custom_model_providers. The endpoint is unreleased, so no published client is affected.
20.6.0 (2026-09-03) Features app: add Z.ai as a built-in GLM model provider (58b7017) js: reorganize assistant settings page into topical tabs (c41b8b8) playground: add support for Gemini 3.8 Flash (20c8be3) server: add DELETE /projects/{project_identifier}/traces (bf659ca) Bug Fixes cost: update built-in model token prices (466863d) experiments: count...
3.6.0 (2026-09-01) Features evals: add session-level pii_detection evaluator (0e3b2c2) Bug Fixes evals: use non-blocking sleep in async rate limiter (2ecfd47)
20.4.0 (2026-08-26) ⚠ BREAKING CHANGES server: GET /v1/model_providers no longer returns custom providers or a next_cursor, and built-in entries expose provider rather than kind + provider_key. Custom providers move to GET /v1/custom_model_providers. The endpoint is unreleased, so no published client is affected.
3.3.0 (2026-08-22) Features pxi: enable phoenix-gql mutations by default with approval in manual mode (391d0dd)
3.5.0 (2026-08-21) Features evals: add retrieval relevance evaluator (7efa7b2)
3.2.0 (2026-08-14) Features server: add POST /traces/transfer (5c38cc7)
20.2.0 (2026-08-13) Features agents: emit a machine-readable approval decision on gated PXI tools (f46fced) Bug Fixes cost: update built-in model token prices (36d02ab) Documentation add Phoenix release notes (63e89b0) filters: add span & session filter expressions reference (876c5a0)
20.1.0 (2026-08-12) Features auth: support JWT client assertion for OAuth2 login (3b277cf) refine evaluation metric chart layout and controls (39740fa) server: add POST /traces/transfer (5c38cc7) Bug Fixes agents: prevent exported session ID collisions (4e828c3) route OpenAI reasoning models to the Responses API client (2d44558) Documentation add v20.0.0...
3.0.0 (2026-08-11) ⚠ BREAKING CHANGES client: replace google-generativeai formatter with google-genai Features add REST endpoint for prompt metadata updates (244ef9c) agents: move PXI tracing server-side (ea40aea) api: add REST endpoints for setting tags on experiments (c892873) api: expose dataset example source span (e0cf022) api: REST CRUD for user and...
19.20.0 (2026-08-10) Features add copy actions to session turns (b9999ae) server: add dataset split create/update/delete REST endpoints (b4d9b19) Bug Fixes app: unblock project navigation during loader fetches (e11b1c3) deps: update arize-phoenix-otel to 0.17.0 (f413867)
3.4.0 (2026-08-08) Features evals: add hallucination evaluator (1aa1a84)
19.19.0 (2026-08-06) Features ui: collapse all but the last message in LLM spans (11f418f) Bug Fixes db: replace sqlean.py with arize-phoenix-sqlean (0fc6f0a) Documentation remove Arize AX from tutorial and get-started flows (e75fec0) update API reference to include chat completions (3af38bd)
0.1.0 (2026-08-05) Features phoenix-sqlean: vendor nalgeon/sqlean.py as arize-phoenix-sqlean (3945ff4)
19.16.0 (2026-08-04) Features server: OpenAI-compatible v1/chat/completions proxy with server-side credentials (6ab57db) ui: preview card contents while collapsed (98b82ba) Bug Fixes render exception stack traces in events (5ca716e) ui: carry the time range forward from the project list into project details (d9421a4) Performance Improvements ui: parallelize...
19.14.0 (2026-08-03) Features evals: add hallucination evaluator (1aa1a84) Bug Fixes traces: render ReasoningPart in gen_ai message flattening (05d5999), closes #14962
19.13.0 (2026-08-01) Features app: include span and trace annotations in span exports (f9fa685) experiments: lift metric chart selection limit (6f647cb)
19.12.0 (2026-07-31) Features evals: harbor poc (24ff59b) Bug Fixes dsl: one NUL message on every Python (1717c4c) Documentation Add Phoenix release notes , 2026-07-29 (ae690e1)
19.11.0 (2026-07-30) Features add project evaluation metrics charts (7ef4832) evals: count subagent tool calls online (819c09b) project: defer metric chart loading and lift the chart selection cap (b455a3f) sandboxes: use Pydantic logo for Monty provider icon (73abdaf) show tool and tool call counts in LLM span card headers (0be0c11), closes #14712 trace...
19.10.0 (2026-07-28) Features add experiment evaluation metrics charts (661cf9e) breakdown token charts by sub-type (4e161f7) charts: reorder selected charts by dragging in the chart menu (3599c77), closes #14793 experiments: collapse experiment table rows to a single line (e0c97cd) sandbox: add Monty provider with shared runtime (342b229) show cost and...
19.7.0 (2026-07-27) Features add span detail downloads (a61a86c) dsl: add parent_span is None root-span predicate to span filter DSL (89e54b5) dsl: analyze root-span scope, stop applying it twice (aa9254e) pytest: evaluator trace isolation and experiment metadata for the pytest plugin (d81ee21) ui: add SegmentedControl component (649b061) ui...
19.5.0 (2026-07-23) Features add evaluation metrics aggregation APIs (ad41a4d) agents: online trace evals with tool_count_per_turn and user_friction (bdff4b5) api: expose dataset example source span (e0cf022) evals: add toxicity gallery template with input/output-agnostic benchmark (1082155) playground: add Gemini 3.6 Flash and 3.5 Flash-lite support...
3.3.0 (2026-07-23) Features evals: add toxicity gallery template with input/output-agnostic benchmark (1082155)
19.4.0 (2026-07-22) Features components: align span kind icon and timeline bar colors with token treatment (51cf276) components: make IDBadge the consolidated click-to-copy ID element (9445683) js: upgrade phoenix-otel to openinference-vercel v3 for AI SDK v7 spans (30f0827) tracing: redesign span header into identity and meta rows (b1a6359) Bug Fixes...
3.2.0 (2026-07-21) Features evals: add user friction evaluator (1ae1a39) Bug Fixes evals: improve user friction prompt (115d3f2) evals: skip LiteLLM on Python 3.14 (bc9b533) Documentation fill small feature-gap docs across evals, tracing, playground, and packages (e5764b3)
19.3.0 (2026-07-20) Features auth: negotiate HTTP/2 for OAuth2/OIDC provider requests (1bf1603) playground: filter model picker to provisioned providers (2f14ce8) settings: add documentation onramps (2e748ef) ui: add routed profile settings tabs (326ed65) Bug Fixes agents: omit trace metadata for PXI turns when tracing is off (3f3d853) evals: improve user...
19.2.0 (2026-07-19) Features cli: improve server startup banner (3904c49) server: show dev tooling info in boot banner (7e36376) settings: add Antigravity and OpenCode MCP client instructions (8180357) Bug Fixes app: inline codemirror-json-schema in vitest to fix ESM resolution (dc0d641) cli: sanitize startup banner on windows (1474c9b) playground: assume...
19.1.0 (2026-07-18) Features cli: px setup mcp , register the Phoenix remote MCP server with a coding agent (f9cd97a) ui: add icon prefixes to settings tabs (7868a58) ui: add MCP settings tab with setup guide (0587fca) Documentation rewrite agent-assisted setup around px setup; MCP docs updates (f3d7af3)
19.0.0 (2026-07-17) ⚠ BREAKING CHANGES MIGRATION.md - auth: GraphQL createUserApiKey and createSystemApiKey no longer accept API-key-authenticated callers. Workflows that used an existing API key for unattended key creation must issue keys from a session-authenticated context or, for system keys, PHOENIX_ADMIN_SECRET.
18.0.0 (2026-07-14) ⚠ BREAKING CHANGES MIGRATION.md sessions: use interval-overlap semantics for session time-range filters Features sessions: use interval-overlap semantics for session time-range filters (cd17357)
17.28.0 (2026-07-13) Features ui: add session annotation editing in session details (522a9a9) ui: collapsible playground sections via TitledPanel (2858f8f) ui: modernize core Tabs with animated selection indicator and scroll snapping (93016bf) Bug Fixes harden one-click deploy templates (1466235) ui: propagate IO tooltip cell clicks (f675b2a) Documentation...
2.13.0 (2026-07-12) Features client: add name_contains filter to Python client projects.list() (fa79dc0), closes #14034 phoenix env file (c0ab6a9) sdk: accept both PHOENIX_PROJECT_NAME and PHOENIX_PROJECT everywhere (1e7d9fc)
17.25.0 (2026-07-11) Features agents: approval-gated annotation config tools + span-coding for PXI (189271b) phoenix env file (c0ab6a9) ui: add Metrics tab to the dataset page with experiment charts (2c00ee5) ui: refine experiment metric charts (f1eb3b8) Bug Fixes ui: keep filters applied on project tables during live streaming refresh (fd6ce19) ui: make...
17.24.0 (2026-07-10) Features evals: move PXI evals to pytest plugin (2d5052c) Bug Fixes agents: retry hallucinated native tool calls (a6f867e) cost: update built-in model token prices (6f16e4e)
17.22.0 (2026-07-09) Features ui: improve PXI onboarding gate UX (16e5638) Bug Fixes agent: let Command+I close the assistant popover while chat input is focused (3977b36) agents: raise Anthropic output token cap (029b6f6) app: scroll settings tab bar instead of wrapping on narrow widths (1faefa7) ui: remove permanent scrollbar on trace tree panel (8f76524)...
17.21.0 (2026-07-08) Features agents: end-to-end PXI turn tracing from browser to backend (80dda23) app: global search command palette (⌘K) (3a8061b) read only styles (63fcad2) session stats side panel for the sessions table (84d18bd) session turn divider with copy (546534a) sessions: sweep orphaned sessions after a one-hour grace period (8d699e7) ui: show...
17.20.0 (2026-07-07) Features ui: user-selected metric charts above project tables (54f50e2) Bug Fixes agents: improve chat error recovery (c8bf700) cost: update built-in model token prices (7d19cfe) ui: guard against nullish projects connection on Projects page (7a88302) Documentation Add cookbook on "Identifying high-signal traces in production" (184811c)...
Releasesby Glyph