From b124e0203ebf67382191eb9779319c7fe2a9a6c5 Mon Sep 17 00:00:00 2001
From: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Date: Thu, 11 Jun 2026 19:15:57 +0000
Subject: [PATCH] ci: rightsize llm_translation job and fold responses unit
 survivors in

Per the keep/drop audit (8d):

- llm_translation_testing: resource_class xlarge -> large. After the drop and
  harness migration the job is ~17k lines of mostly deterministic unit tests
- remove --retries 2 --retry-delay 5: retries existed to mask live-provider
  flake; the surviving tests are deterministic, so a pass-on-retry should be
  treated as a bug, not absorbed. -n 4 and --timeout=120 stay
- llm_responses_api_testing job deleted; its glob folds into
  llm_translation_testing since only ~2.4k lines of unit survivors remain
  (live Responses tests now live in tests/harness_suites/responses_api_live)
- per-commit recorder jobs (e2e_openai_endpoints,
  proxy_e2e_anthropic_messages_tests, proxy_spend_accuracy_tests) untouched,
  as are realtime/agent/guardrails/google_generate_content/ocr jobs (out of
  chat scope)

Provider API keys stay in the job context for now: the non-chat live tests
listed in tests/harness_suites/HANDOFF.md still run here until Sameer's
suites take them.
---
 .circleci/config.yml | 50 +++++---------------------------------------
 1 file changed, 5 insertions(+), 45 deletions(-)

diff --git a/.circleci/config.yml b/.circleci/config.yml
index dbeb412506f..09250167523 100644
--- a/.circleci/config.yml
+++ b/.circleci/config.yml
@@ -722,7 +722,7 @@ jobs:
     docker:
       - *python312_image
     working_directory: ~/project
-    resource_class: xlarge
+    resource_class: large
 
     steps:
       - checkout
@@ -747,8 +747,10 @@ jobs:
             # Add --durations=20 to show 20 slowest tests for debugging
             # Subdirectories with dedicated jobs (maintain this list as new jobs are added)
             mkdir -p test-results
-            # Glob excludes the realtime/ subdirectory since it has its own job
-            TEST_FILES=$(circleci tests glob "tests/llm_translation/**/test_*.py" | grep -v "^tests/llm_translation/realtime/")
+            # Glob excludes the realtime/ subdirectory since it has its own job.
+            # llm_responses_api unit survivors fold in here since their live
+            # tests moved to the 4h harness (tests/harness_suites/).
+            TEST_FILES=$(circleci tests glob "tests/llm_translation/**/test_*.py" "tests/llm_responses_api_testing/**/test_*.py" | grep -v "^tests/llm_translation/realtime/")
             echo "$TEST_FILES" | circleci tests run \
               --verbose \
               --command="tr ' ' '\\n' | awk '/\\.py/ {print; next} {sub(/\\.[A-Z][^.]*$/, \"\"); gsub(/\\./, \"/\"); print \$0 \".py\"}' | xargs uv run --no-sync python -m pytest \
@@ -757,7 +759,6 @@ jobs:
                 --durations=20 \
                 -n 4 \
                 --timeout=120 --timeout_method=thread \
-                --retries 2 --retry-delay 5 \
                 --max-worker-restart=5"
           no_output_timeout: 15m
 
@@ -938,45 +939,6 @@ jobs:
             - google_generate_content_endpoint_coverage.xml
             - google_generate_content_endpoint_coverage
 
-  llm_responses_api_testing:
-    docker:
-      - *python312_image
-    working_directory: ~/project
-    resource_class: large
-
-    steps:
-      - checkout
-      - setup_google_dns
-      - install_uv
-      - restore_cache:
-          keys:
-            - v1-uv-cache-{{ checksum "uv.lock" }}
-      - run:
-          name: Install Dependencies
-          command: |
-            uv sync --frozen --all-groups --all-extras --python 3.12
-      - save_cache:
-          paths:
-            - ~/.cache/uv
-          key: v1-uv-cache-{{ checksum "uv.lock" }}
-      # Run pytest and generate JUnit XML report
-      - run:
-          name: Run tests
-          command: |
-            mkdir -p test-results
-            TEST_FILES=$(circleci tests glob "tests/llm_responses_api_testing/**/test_*.py")
-            echo "$TEST_FILES" | circleci tests run \
-              --verbose \
-              --command="tr ' ' '\\n' | awk '/\\.py/ {print; next} {sub(/\\.[A-Z][^.]*$/, \"\"); gsub(/\\./, \"/\"); print \$0 \".py\"}' | xargs uv run --no-sync python -m pytest \
-                -v -x \
-                --junitxml=test-results/junit.xml \
-                --durations=5 \
-                -n 8"
-          no_output_timeout: 15m
-
-      # Store test results
-      - store_test_results:
-          path: test-results
   ocr_testing:
     docker:
       - *python312_image
@@ -2957,8 +2919,6 @@ workflows:
           filters: *main_branches
       - google_generate_content_endpoint_testing:
           filters: *main_branches
-      - llm_responses_api_testing:
-          filters: *main_branches
       - ocr_testing:
           filters: *main_branches
       - search_testing: