Merge pull request #45930 from BerriAI/litellm_ui_bundle_rc_1_106_0

chore: rebuild Admin UI bundle for rc/1.106.0

Merge pull request #45927 from BerriAI/litellm_ui_bundle_rc_1_105_0

chore: rebuild Admin UI bundle for rc/1.105.0

Merge pull request #45936 from BerriAI/litellm_guardrail_settings_imp…

…ort_1104

fix(guardrails): import the decisions provider list from litellm.decisions.call on stable/1.104.x

feat(ui): configure OpenAI workload identity federation from the LLM …

…Credentials and Add Model forms (#45528)

* feat(ui): configure OpenAI workload identity federation from the LLM Credentials and Add Model forms

* fix(ui): let a federated OpenAI credential omit the service account when the proxy env provides it

* fix(ui): clear the federation API Base error when the admin switches the credential to an API key

chore: bump litellm-proxy-extras to 0.4.107 and litellm-enterprise to…

… 0.1.75 (#45291)

Co-authored-by: yuneng <yuneng@berri.ai>

feat(decisions): backport /v1/systemone, OpenAI-format /v1/decisions …

…and the openai Decisions provider to rc/1.105.0 (#44236, #45184, #45214) (#45189)

* feat(decisions): add unified /v1/decisions endpoint for Jev-compatible providers (#44236)

Backport of #44236 to rc/1.105.0.
Cherry-picked from 8b1990b (main).

* feat(decisions): serve System One format at /v1/systemone and OpenAI format at /v1/decisions

Backport of #45184 to rc/1.105.0.
Cherry-picked from 3339469 (litellm_systemone_endpoint). The playground System One tab is not on this line, so its UI hunks are dropped.

* fix(decisions): import assert_never from typing_extensions for Python 3.10

Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from 29e7154 (litellm_systemone_endpoint).

* fix(decisions): cap OpenAI-format questions at the System One limit

Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from 7ef1329 (litellm_systemone_endpoint).

* fix(decisions): return 400 for bodies that are not JSON

Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from b6bb885 (litellm_systemone_endpoint).

* test(decisions): send System One bodies to /v1/systemone in integration tests

Backport of the #45184 follow-up to rc/1.105.0.
Cherry-picked from 77872f6 (litellm_systemone_endpoint).

* feat(decisions): add OpenAI as a Decisions provider

System One requests to openai/ models are translated to OpenAI's Decisions shape on the way out and OpenAI's answers are translated back, so both /v1/systemone and /v1/decisions can route to gpt-6-luna.

(cherry picked from commit b94bbbc)

* refactor(decisions): move the OpenAI wire translation into llms/openai

(cherry picked from commit d209a85)

* feat(decisions): route both request formats through a shared decisions IR

System One and OpenAI-format bodies now convert to one internal representation, and each provider translates it to its own wire format. OpenAI deployments receive the caller's messages, images, names, typed choice values, level descriptions and safety_identifier unchanged. System One providers return a 400 for image input. Cached and cache-write tokens are priced on both response shapes, and provider response extras survive the round trip.

(cherry picked from commit 0dd207f)

* fix(decisions): keep provider fields inside System One answers

(cherry picked from commit 1348de8)

* fix(decisions): load on Python 3.10 and 3.11 and bill cached tokens once under custom pricing

Decisions IR dataclasses used a mappingproxy default, which is only hashable from Python 3.12, so import litellm failed on 3.10 and 3.11. Decisions usage now reaches cost_per_token as a Usage with prompt_tokens_details, so custom pricing no longer adds cached and cache-write tokens on top of input_tokens that already include them

(cherry picked from commit 39012f1)

* fix(decisions): honor litellm OpenAI key and base settings for the openai provider

OpenAI decisions now resolve the key from litellm.api_key, litellm.openai_key, then OPENAI_API_KEY, and the base from litellm.api_base, OPENAI_BASE_URL, then OPENAI_API_BASE, matching other OpenAI calls. Providers own their configured key and base lookup through the endpoint config.

(cherry picked from commit e465919)

* test(decisions): clear every OpenAI base setting in the proxy OpenAI deployment test

(cherry picked from commit a2951b3)

* fix(decisions): send OpenAI instructions for System One questions without them and reject single-option questions

(cherry picked from commit e4081f7)

* fix(decisions): return guardrail_information on OpenAI-format decisions when requested

Backport of the #45184 and #45214 guardrail fix.
Cherry-picked from 855f0bb (litellm_systemone_endpoint) as merged into #45214 at c19ada6 (litellm_openai_decisions_provider).

---------

Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>

Merge pull request #45190 from BerriAI/litellm_backport_systemone_sta…

…ble_1_104_x

feat(decisions): backport /v1/systemone, OpenAI-format /v1/decisions and the OpenAI Decisions provider to stable/1.104.x for v1.104.2 (#44236, #45184, #45214)

Merge pull request #45215 from BerriAI/litellm_backport_1_102_x_bp44508

chore(release): backport #44508 to stable/1.102.x and cut 1.102.4

Merge pull request #45212 from BerriAI/litellm_backport_1_101_x_bp40989

chore(release): backport #40989 to stable/1.101.x and cut 1.101.6

ci: run integration-mcp on an xlarge machine (#44608)

Co-authored-by: yuneng <yuneng@berri.ai>