litellm/tests/integration/routing
Mateo Wang 6cdb9959df
feat(router): opt in to non-chat deployment filtering (#46257)
* feat(proxy): reject chat routes on models declared non-chat via model_info.supported_endpoints

A model whose every deployment sets a non-chat model_info.mode and a supported_endpoints list with no chat endpoint now answers /v1/chat/completions, /v1/messages and /v1/responses with a 400 before any provider call. Models without both fields keep today's behavior.

* fix(proxy): let fallbacks answer and cover more chat routes in the non-chat model check

The check now steps aside when a request, key, team or router fallback would answer, skips such models in the rate-limit fallback loop, ignores mode and supported_endpoints copied from the cost map by old Admin UI saves, treats a blank mode as unset, and also covers /v1/completions, the Responses WebSocket, Gemini generateContent and /v1beta/interactions

* fix(proxy): only step aside for fallbacks that can answer chat, gate the Responses WebSocket after alias resolution, and compare declarations to the loaded catalog

* fix(proxy): gate the Responses WebSocket with the router after aliases resolve, and ignore client-set internal fallback flags

* fix(proxy): walk fallback groups breadth first and release reservations on a rejected WebSocket

Each fallback group is looked up once however often the request repeats it, and a WebSocket closed by the endpoint gate runs the failure hook so rate-limit reservations are released

* fix(proxy): walk fallback layers without recursion and gate the compact Responses route

The fallback walk now folds one layer at a time, so the recursion check passes, and /v1/responses/compact on a model declared non-chat gets the same 400 as the other chat-shaped routes

* fix(proxy): keep chat on rows whose cost map entry serves chat

A row that carries the cost map key of an entry that serves chat counts as declaring nothing, so a catalog edit that adds a chat path never turns old Admin UI copies into 400s

* fix(proxy): check chat sent to a deployment id or backend model name

* fix(proxy): check a model the read-through loads from the database

* fix(proxy): step aside for fallbacks the check cannot follow and wildcards that serve a backend name

A fallback target with no deployments of its own, a client routing stamp, and a wildcard that does not match the requested backend name no longer turn a working request into a 400

* fix(proxy): ignore a client model_group that is not a string when building rate-limit headers

A client metadata.model_group sent as an int or a list crashed the error handler with a 500 once the embedding-only check rejected the call before routing. The header helper now treats a non-string stamp as no model group. Drops the pattern clause in the backend-name lookup, since the group lookup already returns wildcard matches. Adds the third audit round's routing-stamp cells.

* fix(proxy): keep a non-string client model_group out of spend rows

* fix(proxy): validate standard fallbacks before the typed fallback-group lookup

* feat(router): opt in to non-chat deployment filtering

* fix(router): skip unresolved local rate-limit fallbacks

* fix(router): ignore caller-provided endpoint selector values

* fix(router): normalize declared chat endpoint prefixes

* test(router): cover preflight error-value behavior

* test(router): derive replay metadata defaults from model info

* fix(router): preserve fallback aliases and check synchronous text calls

* test(routing): rewrite the WebSocket model with a pre-call callback, since main no longer runs guardrails at socket admission

* test(routing): keep the endpoint gate's owned proxies on the worker's database

* fix(router): keep the endpoint gate on model_info patches that omit it and on same-group stream retries

A partial PATCH of model_info reset a stored enforce_supported_endpoints true back to false, because the
field defaulted to False and the update merges model_dump(exclude_none=True). The field now defaults to None
so an omitted flag keeps the stored value

The same-group retry of a stream dropped before content called the provider without the endpoint selection
context, so it could land on an enforced embedding-only deployment of the same group. It now runs inside
endpoint_selection like every other router attempt

* fix(router): probe rate-limit fallbacks through get_routable_deployments so the router surface does not grow

Main's router fence lets code outside the router use only listed Router members and
router_utils modules. The proxy's fallback probe now calls the listed
get_routable_deployments with the request's call type, and the gate's 400 moves to
litellm.exceptions as EndpointNotSupportedByModelError. It stays separate from main's
UnsupportedEndpointError, which the router retries on the group's other deployments,
while the gate refuses the whole group

* test(responses): cover the WebSocket error frame for a model that refuses the endpoint

* fix(proxy): let the rate-limit fallback probe try a fallback this worker has not loaded yet

The probe now skips a fallback only when the endpoint check removed every deployment it has, so a model added on another worker still reaches route_request and its database lookup

* fix(proxy): keep key router settings off the Responses WebSocket, read unrooted or single-string endpoint lists, and add status to the refusal frame

* fix(router): warn when enforce_supported_endpoints has no usable list and match chat endpoints regardless of case, trailing slash, or Gemini path

* test(router): type the load-warning test's caplog and router

* test(routing): match the endpoint gate's integration tests to base fallback behavior on unloaded fallbacks and the Responses WebSocket

---------

Co-authored-by: Mateo <mateo@Mateos-MBP-2.localdomain>
2026-10-29 08:33:36 -07:00
..
_prompt_cache_pin_support.py fix(router): keep the prompt caching affinity pin alive as long as the prefix's cache_control ttl (#46084) 2026-10-14 21:10:09 -07:00
_reasoning_effort_levels_support.py test(routing): audit #46601 (#48910) 2026-10-28 15:42:51 -07:00
either_role.json test(integration): add read-replica routing harness to the CircleCI integration suite (#42692) 2026-09-24 00:25:03 -07:00
test_advisor_failure_cooldown.py test(integration): regression tests for July provider translation, routing and streaming bugs (#42693) 2026-09-23 09:51:06 -07:00
test_audit_any_sweep_chaos.py test(integration): give chaos and timing tests a harness API and gate raw signals (#46031) 2026-10-12 05:27:33 -07:00
test_auto_router_management.py test(integration): wait on the right signal in five flaky integration contracts (#46737) 2026-10-15 13:48:51 -07:00
test_azure_passthrough_spend_row.py fix(vllm): write spend log rows for /vllm passthrough calls to router models (#45733) 2026-10-14 12:40:48 -07:00
test_base_model_context_window_owned_proxy.py fix(router): let an exact base_model widen a deployment's context window (#46197) 2026-10-16 06:37:40 +00:00
test_batch_mode_deployments_wire.py test(router): audit #47376 (#48792) 2026-10-28 04:48:30 -07:00
test_complexity_router_codex_compaction.py fix(complexity_router): keep the held user_turn route for Codex compaction requests (#48627) 2026-10-27 13:33:32 -07:00
test_complexity_router_databricks_classifier.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_complexity_router_llm_classifier.py refactor(types): replace Any with proven types in 7 files (#43844) 2026-10-02 02:47:47 -07:00
test_declared_provider_bare_model_routes.py test(integration): hand every harness server a port from one per-worker block (#46761) 2026-10-15 16:56:40 -07:00
test_deployment_mode_routing_wire.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_deployment_num_retries_generic_routes_wire.py fix(router): retry a /v1/messages stream the provider drops before the first content chunk (#44276) 2026-10-05 22:53:39 +00:00
test_encrypted_content_affinity_no_peer.py fix(router): serve Responses turns from a sibling when the encrypted content origin has no boundary peer (#43015) 2026-09-24 17:31:43 -07:00
test_end_user_region_routing.py fix(router): drop the deployment rpm filter from enable_pre_call_checks (#48873) 2026-10-28 12:13:56 -07:00
test_enforce_model_rate_limits_daily_limits.py feat(router): add deployment rpd and tpd daily limits to enforce_model_rate_limits (#48875) 2026-10-28 15:52:44 -07:00
test_fallback_hop_foreign_encrypted_reasoning.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_fallback_retry_header_contracts.py test: move 76 live top-level tests to offline unit and integration coverage (#45352) 2026-10-08 10:33:35 -07:00
test_include_fallback_errors_wire.py fix(router): keep include_fallback_errors off the provider call on the sync Router path (#45218) 2026-10-07 22:25:12 -07:00
test_key_max_parallel_requests.py test(integration): stop GET /__observations from draining the upstream log (#47256) 2026-10-18 09:58:05 +00:00
test_key_tpm_reservation.py test(integration): regression tests for July provider translation, routing and streaming bugs (#42693) 2026-09-23 09:51:06 -07:00
test_messages_stream_retry_budget_sources_owned_proxy.py test(integration): fix seven CircleCI integration tests that race or assume a worker spread (#47001) 2026-10-16 22:45:29 +00:00
test_model_capability_flags.py fix: keep model capability flags out of provider requests (#46647) 2026-10-15 14:30:10 +00:00
test_non_chat_model_endpoint_gate.py feat(router): opt in to non-chat deployment filtering (#46257) 2026-10-29 08:33:36 -07:00
test_observed_routing.py test(integration): stop GET /__observations from draining the upstream log (#47256) 2026-10-18 09:58:05 +00:00
test_payment_required_cooldown.py fix(router): keep a lone deployment warm under router allowed_fails unless the policy names the exception (#48644) 2026-10-27 15:12:53 -07:00
test_payment_required_mapping.py fix(exceptions): map upstream 402 to PaymentRequiredError and cool down 402 deployments (#44879) 2026-10-06 20:35:13 -07:00
test_pdf_document_pre_call_checks_owned_proxy.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_per_model_rate_limit_fallbacks.py test(integration): give chaos and timing tests a harness API and gate raw signals (#46031) 2026-10-12 05:27:33 -07:00
test_per_model_rate_limit_fallbacks_lifecycle.py test(integration): give chaos and timing tests a harness API and gate raw signals (#46031) 2026-10-12 05:27:33 -07:00
test_pre_call_response_format_filter.py test(integration): stop GET /__observations from draining the upstream log (#47256) 2026-10-18 09:58:05 +00:00
test_priority_model_tpm_enforcement.py test(integration): regression tests for July provider translation, routing and streaming bugs (#42693) 2026-09-23 09:51:06 -07:00
test_priority_rate_limit_headers.py chore(lint): remove the LIT002 mutable-construction rule (#43971) 2026-10-01 12:24:02 -07:00
test_priority_scheduler_queue_cleanup.py test(router): audit #47498 (#48799) 2026-10-28 04:36:50 -07:00
test_prompt_cache_pin_memory_routes.py fix(router): keep the prompt caching affinity pin alive as long as the prefix's cache_control ttl (#46084) 2026-10-14 21:10:09 -07:00
test_prompt_cache_pin_ttl.py test(integration): repair seven stale integration tests on main (#47600) 2026-10-20 11:21:04 -07:00
test_prompt_cache_pin_ttl_chaos.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_prompt_caching_affinity_tool_call_marks.py fix(caching): count tool_call cache_control marks in the injection census (#43556) 2026-10-03 17:33:48 -07:00
test_ptu_shares.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_ptu_shares_regex_timing.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_reasoning_effort_levels_chaos.py feat(router)!: apply each deployment's reasoning_effort_levels on every provider and route (#46601) 2026-10-20 04:34:55 +00:00
test_reasoning_effort_levels_wire.py test(routing): audit #46601 (#48910) 2026-10-28 15:42:51 -07:00
test_redis_increment_with_floor.py ci(circleci): test Redis behavior against local Redis and print short tracebacks (#44062) 2026-10-01 13:00:54 -07:00
test_redis_recovery.py test(integration): stop GET /__observations from draining the upstream log (#47256) 2026-10-18 09:58:05 +00:00
test_responses_websocket_turn_admission.py fix(tests): read owned proxy rows from the owned proxy's database (#47333) 2026-10-20 10:07:06 -07:00
test_router_strategies_count_async_requests_once.py fix(logging)!: fire one CustomLogger failure hook per async call outside the async flag list (#47124) 2026-10-20 15:19:14 -07:00
test_silent_model_shadow_wire.py test(router): audit #41368 (#48794) 2026-10-28 05:14:59 -07:00
test_stale_cost_map_boot.py test(integration): stop GET /__observations from draining the upstream log (#47256) 2026-10-18 09:58:05 +00:00
test_tag_routing_prefix.py fix(router)!: only prefixed caller tags route once tag_routing_prefix is set (#47428) 2026-10-20 00:12:28 -07:00
test_team_tag_routing.py test(integration): stop GET /__observations from draining the upstream log (#47256) 2026-10-18 09:58:05 +00:00
test_team_wildcard_precedence.py test(routing): a team's wildcard serves its keys before a proxy-wide wildcard of the same pattern (#48501) 2026-10-26 17:54:32 -07:00
test_usage_based_routing_redis_reads.py test: fix stale and flaky tests across CircleCI, GHA and Buildkite (#45530) 2026-10-09 11:40:14 -07:00
test_user_config_routing.py feat(openai)!: honor per-deployment workload identity on non-chat routes (#39966) 2026-10-16 22:25:15 +00:00
test_video_poll_routing_wire.py fix(videos)!: keep model_id in proxy video ids so per-model keys resolve on status/content (#44614) 2026-10-20 05:42:59 +00:00
test_vllm_passthrough_spend_row.py fix(vllm): write spend log rows for /vllm passthrough calls to router models (#45733) 2026-10-14 12:40:48 -07:00
test_vllm_passthrough_stream_relay.py fix(vllm): write spend log rows for /vllm passthrough calls to router models (#45733) 2026-10-14 12:40:48 -07:00
test_vllm_passthrough_stream_usage_settings.py test(integration): give each owned proxy a private database and Redis by default (#47213) 2026-10-18 03:37:41 +00:00
test_wildcard_model_listing.py fix(providers): route Azure Responses past dated api versions, count Grok reasoning tokens on azure_ai, refresh provider sets after register_model, fix gpt-6.1-sol Azure rows (#46377) 2026-10-16 15:23:46 -07:00