litellm/tests
Rocky Li 51553b4621 fix(anthropic-adapter): correct initial content block type and preserve first delta on block transitions
When models return reasoning_content (e.g. GLM-5 via Vertex AI), the
Anthropic pass-through adapter now:

1. Peeks at the first streaming chunk to determine the correct initial
   content_block_start type (thinking vs text) instead of hardcoding text.

2. Detects reasoning_content (without thinking_blocks) as a thinking
   block type in _translate_streaming_openai_chunk_to_anthropic_content_block.

3. Queues the processed chunk during content block transitions so the
   first token of the new block is not silently dropped.

These fixes ensure Anthropic API consumers (e.g. Claude Code CLI) receive
a valid SSE event sequence where thinking_delta events arrive inside a
thinking content block, not a text block.

Made-with: Cursor
2026-04-06 15:31:37 +08:00
..
agent_tests refactor: make unit test 2026-03-28 21:44:46 -07:00
audio_tests test: cleanup test impacted by deployment name change 2026-03-30 18:07:50 -07:00
basic_proxy_startup_tests CircleCI test stability (#23055) 2026-03-07 15:19:39 -08:00
batches_tests test fix us.anthropic.claude-haiku-4-5-20251001-v1:0 (#24931) 2026-04-01 11:01:03 -07:00
benchmarks Add CodSpeed performance benchmarks (#23676) 2026-03-14 18:44:36 -07:00
code_coverage_tests [Fix] Remove unused aioboto3 dependency and botocore conflict workarounds 2026-04-01 14:25:44 -07:00
documentation_tests [Fix] Scope documentation workflow to match CircleCI and add missing router settings 2026-03-28 11:23:53 -07:00
enterprise test fix 2026-03-30 18:08:57 -07:00
guardrails_tests Add Akto Guardrails to LiteLLM (#23250) 2026-03-17 14:38:04 -07:00
image_gen_tests fix: cleanup tests 2026-03-30 16:24:35 -07:00
litellm Litellm ishaan april2 (#25113) 2026-04-04 12:31:49 -07:00
litellm-proxy-extras [Infra] Mirror litellm_table_patch source changes (no binaries) 2026-03-27 16:45:12 -07:00
litellm_core_utils Fix: empty assistant message for converse API 2026-02-04 09:50:58 +05:30
litellm_utils_tests Litellm ishaan april4 2 (#25150) 2026-04-04 23:09:42 +00:00
llm_responses_api_testing test: remove bad e2e tests - e2e failing due to low rate limits on ci/cd 2026-03-30 19:18:10 -07:00
llm_translation [Infra] Merge internal dev branch with main (#25036) 2026-04-02 16:38:01 -07:00
load_tests test: update to new vertex ai keys 2026-03-28 20:19:05 -07:00
local_testing Litellm fix update bedrock models (#24947) 2026-04-01 19:22:54 -07:00
logging_callback_tests [Test Fix] fix gov pricing tests (#25022) 2026-04-02 15:55:55 -07:00
mcp_tests [Infra] Merging RC Branch with Main (#23786) 2026-03-16 15:32:20 -07:00
multi_instance_e2e_tests
ocr_tests test: update to new vertex ai keys 2026-03-28 20:19:05 -07:00
old_proxy_tests/tests fix: cleanup tests 2026-03-30 16:24:35 -07:00
openai_endpoints_tests Litellm fix update bedrock models (#24947) 2026-04-01 19:22:54 -07:00
otel_tests fix(tests): increase MAX_CALLS and reduce sleep in flaky e2e budget test 2026-03-13 00:04:31 -07:00
pass_through_tests Litellm fix update bedrock models (#24947) 2026-04-01 19:22:54 -07:00
pass_through_unit_tests Litellm fix update bedrock models (#24947) 2026-04-01 19:22:54 -07:00
proxy_admin_ui_tests [Fix] Remove neon CLI dependency and pin all JS dependencies 2026-04-01 16:15:32 -07:00
proxy_e2e_anthropic_messages_tests test fix us.anthropic.claude-haiku-4-5-20251001-v1:0 (#24931) 2026-04-01 11:01:03 -07:00
proxy_e2e_azure_batches_tests Increase file deletion retry budget to 50s for batch_processed race 2026-03-15 15:39:35 -07:00
proxy_security_tests
proxy_unit_tests cherry-pick: tag query fix + MCP metadata support (#25145) 2026-04-04 16:44:02 -07:00
router_unit_tests fix(router): address remaining Greptile review comments 2026-03-27 20:11:28 +05:30
scim_tests
search_tests Merge remote main, resolve conflict keeping new unit tests 2026-03-09 15:20:20 -07:00
spend_tracking_tests [Staging] - Ishaan March 17th (#23903) 2026-03-18 15:09:01 -07:00
store_model_in_db_tests test: fix assemblyai passthrough transcription request format 2026-03-30 18:49:04 -07:00
test_litellm fix(anthropic-adapter): correct initial content block type and preserve first delta on block transitions 2026-04-06 15:31:37 +08:00
unified_google_tests test: update to new vertex ai keys 2026-03-28 20:19:05 -07:00
vector_store_tests test: update test apis 2026-03-28 18:57:27 -07:00
windows_tests
__init__.py
gettysburg.wav
large_text.py
openai_batch_completions.jsonl
README.MD
test_budget_management.py
test_callbacks_on_proxy.py
test_config.py
test_debug_warning.py
test_default_encoding_non_root.py fix: tiktoken cache nonroot offline (#23498) 2026-03-14 10:48:36 -07:00
test_end_users.py
test_entrypoint.py
test_fallbacks.py Revert "fix: prevent error when max_fallbacks exceeds available models (#20071)" 2026-02-03 15:15:30 +05:30
test_gpt5_azure_temperature_support.py [Fix] Remove deprecated o1-preview from O-series test and deduplicate is_o_series check 2026-03-12 13:22:00 -07:00
test_health.py
test_keys.py
test_litellm_proxy_responses_config.py
test_logging.conf
test_models.py test: replace test_add_and_delete_models integration test with mock 2026-03-30 21:30:57 -07:00
test_new_vector_store_endpoints.py fix(ci): stabilize CI tests - conditional import, mock fixes, timing adjustments 2026-03-13 00:01:25 +00:00
test_openai_endpoints.py
test_organizations.py
test_otel_thread_leak.py Fix thread leak in OpenTelemetry dynamic header path (#19946) 2026-01-28 10:35:37 -08:00
test_passthrough_endpoints.py
test_presidio_latency.py fix(presidio): reuse HTTP connections to prevent OOMs (#19964) 2026-01-28 16:08:53 -08:00
test_proxy_server_non_root.py deactivating non root tests 2026-01-23 22:55:36 -08:00
test_ratelimit.py
test_resource_cleanup.py
test_service_logger_otel.py fix(langfuse_otel): prevent empty proxy request spans from being sent to Langfuse 2026-01-28 15:35:35 +01:00
test_spend_logs.py
test_team.py [Fix] Fail fast when team member spend not flushed in time 2026-03-13 12:40:17 -07:00
test_team_logging.py test: cleanup dead tests 2026-03-28 20:49:02 -07:00
test_team_members.py
test_users.py Litellm fix update bedrock models (#24947) 2026-04-01 19:22:54 -07:00

In total litellm runs 1000+ tests

[02/20/2025] Update:

To make it easier to contribute and map what behavior is tested,

we've started mapping the litellm directory in tests/test_litellm

This folder can only run mock tests.