Backport of #46712 to stable/1.106.x.
Cherry-picked from 9c62573005 (litellm_dashboard_only_gzip). The line never took #46362, so the OFF_LOOP_SIZE_BYTES hunks resolve to the pick, which removes the constant.
Backport adaptation for rc/1.105.0. main merges a pass-through caller's metadata and litellm_metadata into the logging metadata only when they are dicts. This line merged any value, so a list of pairs could carry standard_logging_guardrail_information past the guardrail strip from #59.
Backport of #59 to rc/1.105.0.
Cherry-picked from 79e48b9b19 (main). This line has no client pricing strip on pass-through, so only the guardrail strip is ported there.
metadata and litellm_metadata leave the proxy-only failure-log allowlist, so a rejected request can no longer point its failure log at its own Langsmith or Langfuse host through nested metadata.
(cherry picked from commit 30f6715f480ee7e2854514fe53b91d1b16e1be03)
* fix(deps): upgrade Python dependencies
* fix(deps): lock patched Python dependency versions
* fix(otel): preserve JSON export with the updated SDK
* test(otel): cover JSON export with the updated SDK
* test(otel): update SDK log records and worker discovery
* fix(ui): align Lens peer dependency with patched Next.js
* fix(otel): hand the AgentOps exporter a session it owns for the JWT header
OpenTelemetry 1.45 moved the OTLP HTTP exporter's requests.Session behind a
transport, so writing the minted JWT to the exporter's private _session raised,
was swallowed, and every AgentOps batch went out unauthenticated. The exporter
now creates the session, passes it through the public session= argument and
sets the Authorization header on it
The tests now drive a real export, serving the token endpoint with respx and
capturing the OTLP POST with requests_mock, instead of reading SDK internals
* fix(otel): skip FastAPI instrumentation once the app has started
opentelemetry-instrumentation-fastapi 0.66b1 wraps build_middleware_stack instead
of calling add_middleware, so instrumenting an app whose middleware stack is
already built no longer raises. The call silently did nothing while we still
registered the global response propagator. Check the built stack up front and
skip both, which is what the existing test expects
* test(otel): assert exported requests and span flags instead of SDK internals
OpenTelemetry 1.45 dropped the exporter's private _headers and sets the W3C
random-trace-id flag on new traces, so traceresponse ends in -03. The Langtrace
and Langfuse tests now export through requests_mock and check the URL and headers
the collector receives, and the traceresponse tests expect the server span's own
trace flags
* fix(deps): bump dashboard zod to 4.6.6 and make the vendored Lens use it
zod 4.6.5 is flagged by the dependency scan and blocks osv-scan on every PR. A
root bump alone leaves the vendored Lens tarball, which pins zod 4.6.5 exactly,
on its own nested copy, so a top-level zod override points every consumer at the
root version and the lock carries a single zod 4.6.6
* test(ui): stub the tag summary fetch the client panel retry refreshes
Retrying the model breakdown refetches /tag/summary, which this integration test
left on the real network. When that request failed before the assertion, the
panel switched to its summary error and unmounted the model row, so the test
passed or failed by timing
* fix(deps): drop the duplicated Lens override from the dashboard package.json
* style(ui): format the client panel fetch stub
---------
Dependency files (pyproject.toml, uv.lock and the dashboard package.json and package-lock.json) were left out of this pick and re-derived on the 1.106 line in the following commits
The ClientUsagePanel integration test does not exist on this line, so its hunk was dropped
The traceresponse hunks in test_otel_v2_mount.py were dropped because those tests do not exist on this line
Co-authored-by: Yuneng Jiang <yuneng@berri.ai>
(cherry picked from commit 8f9894fbd6)
Update certifi, Rust trust roots, and the Docker CA bundle to remove emSign ECC Root CA - G3
Require the fixed certifi release in litellm-core and retain libssl3 in Docker runtimes for Prisma compatibility
Co-authored-by: shin-berri <shin-laptop@berri.ai>
(cherry picked from commit f2118fc767)
The bundled beta map stripped dangerous-tool-use-2026-09-03 for providers
main supports it on, so safeguards requests went upstream without the beta
whenever the proxy used the bundled map. The worker-SIGKILL chaos test needs
OwnedProxy.replacing_worker, which this line's integration harness lacks.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
agentcore/search and vertex_ai/semantic-ranker-default-004 are priced in main's map, so these tests follow main: the search test asserts the mapped per-query price and the custom-pricing rerank test uses an unmapped deployment name.
* fix(logging): report pay-as-you-go rates for spilled over PTU requests without base_model
* fix(logging): keep the provider response model so spilled over PTU payloads price the served model
* fix(logging): keep a provider_response_model the response already carries
* fix(logging): read the carried provider_response_model through a typed adapter
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47655 to stable/1.106.x
Cherry-picked from 6b1f4ed46f (main)
* fix(proxy): bill every earlier timed-out retry attempt at its own deployment's model rate
* fix(router): read the retry record's base_model through the typed metadata helper
* fix(router): keep the four-record retry breadcrumb cap
* fix(router): carry a deployment's litellm_params base_model on its retry record
* chore(router): place the cast-ok reason on the cast line and list the base_model helper as covered through log_retry
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47648 to stable/1.106.x
Cherry-picked from 27efc387f7 (main)
* fix(router): price cost-based routing through model_info.base_model
* fix(router): keep cost-based routing pricing helpers fully typed
* fix(router): drop typing.cast from cost-based routing pricing helpers
* fix(router): read the cost map through a typed accessor in cost-based routing
* fix(router): avoid a Final rebind inside the cost-based routing loop
* refactor(router): share deployment_pricing_models across spend, listing and cost-based routing
* refactor(router): keep the /v1/models listing key inline, share pricing models for spend and routing
* refactor(router): inline the deployment pricing-model wrapper in cost-based routing
* refactor(router): read deployment pricing models through one shared helper
deployment_pricing_models(deployment) now owns the model_info.base_model / litellm_params.base_model /
litellm_params.model unpacking, used by cost-based routing and the /v1/models listing. pricing_models(base_model, model)
stays for the failure-row spend estimate, which already holds the two names
* refactor(router): share the str-keyed mapping validator with cost-based routing
* ci: re-run after flaky test_aiohttp_transport read timeout in llms-providers
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47617 to stable/1.106.x
Cherry-picked from 0c6731eb5c (main)
* fix(proxy): price a 499 spend row only for the proxy's own disconnect, and never after the call already completed
* fix(proxy): gate the zero-cost 499 row on the success log being enqueued, not merely claimed
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47652 to stable/1.106.x
Cherry-picked from ffb7133d32 (main)
* fix(proxy): scope the search free-tier counter to the request's provider
* fix(proxy): read the provider with get so partial spend payloads still get the free tier
* test(proxy): give the grounded spend payload fixture its provider
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47646 to stable/1.106.x
Cherry-picked from 640f2d865b (main)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47647 to stable/1.106.x
Cherry-picked from 70ece3501f (main)
* fix(proxy): strip caller base_model from spend pricing and reject lone surrogates in multipart metadata
* fix(proxy): type the parsed multipart metadata so basedpyright does not grow reportArgumentType
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47639 to stable/1.106.x
Cherry-picked from 2a268951c4 (main)
* fix(proxy): bill partial Responses API streams when the client disconnects
Streamed /v1/responses requests and the default-route OpenAI and Azure /v1/messages requests behind them logged a 499 with zero spend when the client went away mid-stream, while /v1/chat/completions already billed the chunks it had sent. The Responses API iterator now keeps the response.created payload and can hand the disconnect cleanup a response.incomplete event with counted usage and the served service_tier, and the messages-route byte stream and router wrappers delegate to it.
* fix(proxy): satisfy lint gates for partial Responses API disconnect billing
Make the disconnect helper public so the proxy, router and messages adapter can import it without reportPrivateUsage, type the SSE byte stream as an AsyncGenerator so aclose is known, read litellm_call_id off the logging object, and keep the new tests on Final locals with the recorder injected through dynamic_async_success_callbacks instead of litellm.callbacks.
* fix(router): bill the primary stream's tokens after a mid-stream hop and chat-bridge Responses streams on disconnect
The router's fallback wrapper now folds the primary Responses stream's relayed tokens into the fallback's disconnect event the way _combine_responses_fallback_usage does for a natural end of stream, and LiteLLMCompletionStreamingIterator (chat-only providers served through /v1/responses) builds its disconnect event from the chat completion chunks it collected. stamp_responses_usage_cost is now public so the bridge module can price that event.
* Potential fix for pull request finding 'CodeQL / Log Injection'
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
Backport of #47611 to stable/1.106.x
Cherry-picked from f0ae445cc4 (main)
Every /v1/responses spend log asked its worker to flush the queue right away so a
chained previous_response_id could find the row. Only the chat-completions bridge
reads that row. Native Responses providers hold the conversation upstream, so their
rows now wait for the batch write like chat rows do
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47614 to stable/1.106.x
Cherry-picked from 423362bab2 (main)
* feat(proxy): opt-in free-tier allowance for Google Search grounding spend
* fix(proxy): declare the search free tier counter's service target
* test(cost_map): allow search_free_requests in the cost map schema
* fix(proxy): skip cache hits and never free more than the billed cost in the search free tier
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47494 to stable/1.106.x
Cherry-picked from e63aec029a (main)
_failure_spend_rows yields the final row plus one re-keyed row per lifted billed_failed_attempts entry (<request_id>-attempt-<position>, the attempt's usage, cost, retry count and deployment id), each written through DBSpendUpdateWriter.update_database so it also deducts from key, team and user spend
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47416 to stable/1.106.x
Cherry-picked from ad8ca533ca (main)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47571 to stable/1.106.x
Cherry-picked from ccd0d4ea49 (main)
* feat(proxy): lift every earlier billed timed-out retry attempt onto the failed request
_billed_earlier_attempts reads the router's previous_models records, keeps the Timeout ones other than the final attempt (matched on deployment_id plus attempted_retries, since the count restarts after a fallback hop and the fallback handler re-logs the exhausting attempt) and prices each at the deployment that served it. post_call_failure_hook lifts the tuple as billed_failed_attempts; nothing writes it yet. previous_models joins the router-reserved metadata fields so a client-planted record never reaches it
* fix(proxy): price earlier timed-out attempts from the prompt estimate only
The final row may lift a mid-stream recovered usage (prompt plus delivered chunks) or a failed_attempts_usage sum; earlier attempts never produced output, so they are priced from _estimated_prompt_usage, and when failed_attempts_usage already bills them into the final row none is lifted again
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47490 to stable/1.106.x
Cherry-picked from 7a4a255d1d (main)
Backport of #47625 to stable/1.106.x.
Cherry-picked from 6564a490f4 (main).
Adapted to the line without #45797 or #47494. The router budget limiter claims the batch charge with the line's duration_in_seconds TTL (budget_window_seconds and the injected clock come from #45797, which is not backported), and its realtime skip keeps the line's existing _aresponses_websocket check. The write_started flag on a budget reservation omits the apply_search_free_tier branch, since that flag is #47494's. Tests: the search-allowance accounting test is dropped for the same reason, and the cost-callback and router tests get the litellm and SimpleNamespace imports the line's test files lack.
Backport of #47419 to stable/1.106.x.
Cherry-picked from 3124797399 (main).
Adapted to the line without #41523. Registered rates are tracked per cost-map key in litellm.utils (_registered_pricing_fields, reset when litellm.model_cost is replaced, so a request-scoped registration does not outlive its catalog) instead of AliasedCostEntry.registered_fields. AliasedCostEntry and get_model_cost_map.py are untouched. The register_model replay now merges successive registrations as on main. An inference-profile ARN model_id is reduced to its profile id inside _invoked_model_id_for_cost rather than backporting the get_model_info ARN handling from #46128. Only #47419's lines are taken in completion_cost; base_model_pricing_provider (#46318) and configured search cost (#46571) are not. Tests: CONVERSE_ANSWER added to the slow_upstream helper (it came with #47162, which is not backported), the isolated_pricing_registry fixture and custom-pricing helpers come from main's test file, and the ignore_stamped_cost assertion uses the line's response_cost_calculator signature.
Backport of #43935 to stable/1.106.x.
Cherry-picked from d7620c77a1 (main).
Adapted to the line: responses_api_bridge_check lives in litellm/main.py here (moved to litellm/responses/bridge_check.py on main by #45772), so the reasoning_active change and _chat_gate_refuses_reasoning_effort_none are ported there, importing AzureAIGPT5Config locally. Cost-map flags are replayed per model key. Test hunks that conflicted are applied onto the line's files.
Prerequisite for the #47190 integration test on stable/1.106.x. Copied unchanged from tests/integration/_support/prisma_clients.py on main, added there by #45939.
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47564 to stable/1.106.x
Cherry-picked from ce63f48ed2 (main)
PR #47394 withheld cache_control_injection_points from the OpenAI SDK extra_body for
every provider on that path, which also covers litellm_proxy. The downstream LiteLLM
proxy applies the point itself, so litellm_proxy keeps forwarding it while OpenAI and the
other OpenAI-compatible providers still never receive it
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47570 to stable/1.106.x
Cherry-picked from 04e9ed4c52 (main)
get_standard_logging_object_payload decided custom_pricing from
litellm_params alone, so on an Azure PTU spillover response
model_map_information kept the deployment's zeroed PTU entry while
response_cost already used the standard model rate. Reuse the billing
side LitellmLogging._custom_pricing_for so both read the same decision
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47567 to stable/1.106.x
Cherry-picked from 4222523261 (main)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47437 to stable/1.106.x
Cherry-picked from fb555b00dc (main)
* fix(proxy): price the estimated input tokens on a client-cancelled (499) request's spend row
* fix(proxy): read the provider handoff before the proxy-only error handler stamps it so a pre-handoff 499 is not priced
* fix(proxy): drop the pre-logging handoff read now that pre_call no longer stamps proxy-only failures
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47436 to stable/1.106.x
Cherry-picked from 0e7c4908bb (main)
* fix(batches): apply the Vertex regional endpoint uplift to batch cost
batch_cost_calculator takes vertex_location and multiplies by the model's
regional_endpoint_uplift_multiplier through get_vertex_regional_endpoint_uplift,
and every batch cost path (aretrieve_batch pricing, completed-batch logging,
native Vertex batch rows, the proxy batch cost poll and its per-line items)
resolves the deployment's location and passes it down
Co-authored-by: Mrun25 <165297479+Mrun25@users.noreply.github.com>
* test(logging): point the resolver tests at resolve_vertex_location_for_cost
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Mrun25 <165297479+Mrun25@users.noreply.github.com>
Backport of #47466 to stable/1.106.x
Cherry-picked from c6c303a618 (main)
Vertex text embedding predict responses report metadata.billableCharacterCount, the
unit Google bills text-embedding-005 and text-multilingual-embedding-002 on. Carry it
into usage.prompt_tokens_details.character_count and, when the model carries
input_cost_per_character, bill that count instead of the token count so the prompt is
charged under one unit. Per-token embeddings (gemini-embedding-001) are unchanged.
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47460 to stable/1.106.x
Cherry-picked from d29200fa05 (main)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47456 to stable/1.106.x
Cherry-picked from 3fa1587f43 (main)
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47435 to stable/1.106.x
Cherry-picked from 7b76b2b15f (main)
* fix(anthropic): apply tool_config cache-control points on chat and /v1/messages
A cache_control_injection_points entry with location tool_config was spread into
the Anthropic request body on /v1/chat/completions, which Anthropic rejected with
400 Extra inputs are not permitted, and the native /v1/messages path dropped the
point silently. Both paths now put the point's control on the last tool Anthropic
caches and never forward the key.
Co-authored-by: Tan Nguyen <tan@pressingly.net>
* refactor(anthropic): validate tool_config inputs instead of casting them
Keeps the lint gates flat: the helper takes untyped tools and points and
narrows them with the existing object validators, the chat transform builds
the body with the marked tools instead of writing into optional_params, and
the handler reads a typed view of kwargs.
* fix(openai): keep tool_config cache points out of extra_body, add integration test
The cache hook hands a tool_config point on for the provider step, and for
providers that call through the OpenAI SDK every non-OpenAI param rides in
extra_body, so OpenAI answered 400 Unrecognized request argument supplied.
The new integration file drives a live proxy against scripted Anthropic,
Bedrock Converse and OpenAI upstreams for /v1/chat/completions and
/v1/messages.
* test(integration): cover tool_config cache points on bridged /v1/messages
Three new cases on /v1/messages for models without a native Anthropic
messages path: OpenAI through the Responses bridge and Gemini through the
chat bridge never see cache_control_injection_points in the body, and
Bedrock Nova through the chat bridge gets the cachePoint on toolConfig
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: Tan Nguyen <tan@pressingly.net>
Backport of #47394 to stable/1.106.x
Cherry-picked from 93b79d2587 (main)
Backport of #47190 to stable/1.106.x.
Cherry-picked from 0ede0e10e7 (main).
Hand-ported onto the line's poller. ManagedObjectRow lives in check_batch_cost.py on this line, so created_at and updated_at are added to its local _ManagedObjectRow. _finalize_unbilled_terminal_job keeps the line's None return and skips the log when the snapshot no longer matches. Test hunks for TestBatchLineItemsFromThePoller are dropped, and the appended tests do not assert the bool return.
Backport of #47182 to stable/1.106.x.
Cherry-picked from 1212260ab1 (main).
Test conflict resolved by keeping the line's file plus this PR's own tests.
Backport of #47184 to stable/1.106.x.
Cherry-picked from 3fb95c8fe3 (main).
Hand-ported onto the line's poller: the claim now writes status and file_object with batch_processed and the release restores them, and _ManagedObjectRow gains status. The line has no postgres_storable_json, so the NUL-stripping sanitizer (_storable_text, _storable_json_value, postgres_storable_json, from #45703 on main) is added to managed_files.py and the claim stores postgres_storable_json(response). Test hunks for TestBatchLineItemsFromThePoller (from #41691, not on the line) are dropped.
* fix(cost): bill Gemini audio/image/video prompt tokens at the resolved input rate when no modality rate exists
Co-authored-by: shivam <shivam@berri.ai>
* style(cost): ruff format the modality cost calls
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: shivam <shivam@berri.ai>
Backport of #47373 to stable/1.106.x
Cherry-picked from 09e9193514 (main)
* feat(router): let a deployment set a timeout per service_tier
Adds timeout_by_service_tier and stream_timeout_by_service_tier to a deployment's
litellm_params. A request whose service_tier matches a key uses that value in place
of the deployment's timeout or stream_timeout. A timeout the client sends still wins
and requests without a matching tier are unchanged
Co-authored-by: adc77 <amandogra2016@gmail.com>
* style(router): sort the request_timeout_resolver import
* fix(router): read the service tier maps without the untyped params get
* fix(router): look the service tier timeout up through a typed get
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Co-authored-by: adc77 <amandogra2016@gmail.com>
Backport of #47406 to stable/1.106.x
Cherry-picked from 62ce515d15 (main)
* fix(decisions): log every request field the caller sent on /v1/systemone and /v1/decisions
* fix(decisions): serialize the logged request with model_dump_json so no mutable dict is built
* refactor(decisions): drop the state-only messages pick in function_setup, prepare_call logs the request
* fix(decisions): return the single logged message so no mutable list type is annotated
* fix(decisions): log the caller's raw request fields before validation so a rejected request is still recorded
* fix(decisions): serialize the raw request fields with pydantic_core so question objects and mapping proxies log as JSON
* fix(decisions): log the caller request at the SDK entry so requests rejected before prepare_call are recorded too
* test(decisions): assert the exact BadRequestError for the rejected-questions case
---------
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Backport of #47183 to stable/1.106.x
Cherry-picked from 53821f6654 (main)
Backport of #43955 to stable/1.106.x.
Cherry-picked from eb5295b20c (main).
Adapted to the line: the existing CLI-beta wire assertion change is dropped since that test is not on the line.
