Skip to content

feat(search): add you_com as a search provider - #28370

Merged
Sameerlite merged 7 commits into
BerriAI:litellm_oss_staging_050626from
brainsparker:you-com-search-provider
Jun 5, 2026
Merged

feat(search): add you_com as a search provider#28370
Sameerlite merged 7 commits into
BerriAI:litellm_oss_staging_050626from
brainsparker:you-com-search-provider

Conversation

@brainsparker

@brainsparker brainsparker commented May 20, 2026

Copy link
Copy Markdown
Contributor

Summary

Registers You.com as a first-class search_provider in the search_tools registry, alongside Tavily, Exa AI, Perplexity, Parallel AI, Brave, Google PSE, DataForSEO, Firecrawl, SearXNG, Linkup, DuckDuckGo, SearchAPI, and Serper.

Once registered, you_com works transparently with the litellm_web_search interception layer across all 25+ supported LLM providers (OpenAI, Anthropic, Vertex, Bedrock, etc.).

Keyless free tier by default. If YOUCOM_API_KEY is not set, the adapter falls through to You.com's keyless endpoint (api.you.com/v1/agents/search, ~100 queries/day, IP-throttled, no signup). This matches the existing keyless-default pattern used by duckduckgo and searxng and means LiteLLM users can try this provider with zero configuration. Setting YOUCOM_API_KEY upgrades to the keyed endpoint (ydc-index.io/v1/search) with higher rate limits.

Refs upstream expansion signal: #15942.

What's in here

New adapter — litellm/llms/you_com/search/transformation.py

  • Endpoint selection:
    • YOUCOM_API_KEY set → POST https://ydc-index.io/v1/search with X-API-Key header
    • no key → POST https://api.you.com/v1/agents/search (keyless free tier)
    • YOUCOM_API_BASE override honored as-is
  • Parameter mapping (Perplexity unified spec → You.com):
    • max_resultscount
    • search_domain_filterinclude_domains
    • countrycountry (lowercased to match Tavily's convention)
    • max_tokens_per_page → not applicable, ignored
  • Response normalization:
    • Flattens results.web + results.news into a single SearchResult list
    • snippet prefers snippets[0], falls back to description
    • page_agedate
  • Both endpoints return an identical JSON shape, verified against live API.

Registry wiring

  • SearchProviders.YOU_COM = "you_com" in litellm/types/utils.py
  • YouComSearchConfig wired into ProviderConfigManager.get_provider_search_config() in litellm/utils.py

Pricing

  • Placeholder you_com/search entry in model_prices_and_context_window.json at 0.0 per query. Happy to update to the public pricing number you'd prefer — let me know.

Tests — tests/search_tests/test_you_com_search.py (7 tests, all pass locally)

  1. test_you_com_search_request_payload — keyed URL, X-API-Key header, count mapping, response normalization
  2. test_you_com_search_domain_filter_and_countryinclude_domains / lowercased country mapping; unified-spec param names don't leak through
  3. test_you_com_search_snippet_fallback_to_description — snippet falls back when snippets is empty
  4. test_you_com_search_news_results_appended — news results flatten in after web
  5. test_you_com_search_complete_url_handles_trailing_slash — normalize trailing slashes on custom api_base
  6. test_you_com_search_keyless_free_tier — keyless URL + absence of X-API-Key when no key configured
  7. test_you_com_search_validate_environment_keyless — config doesn't raise on missing key (keyless is valid)

Usage

Zero-config (keyless free tier):

import litellm
resp = litellm.search(query="latest AI developments", search_provider="you_com")

With API key (higher limits):

import os, litellm
os.environ["YOUCOM_API_KEY"] = "sk-..."
resp = litellm.search(query="latest AI developments", search_provider="you_com", max_results=5)

Router YAML:

search_tools:
  - search_tool_name: "my-you-com-search"
    litellm_params:
      search_provider: "you_com"

Test plan

  • pytest tests/search_tests/test_you_com_search.py -v — 7/7 pass locally
  • tests/search_tests/test_tavily_search.py — passes (adjacent regression sanity)
  • Live verification: both ydc-index.io/v1/search and api.you.com/v1/agents/search return matching JSON shapes
  • CI run on this PR

Registers You.com Search API as a first-class `search_provider` in the
`search_tools` registry, alongside Tavily, Exa, Perplexity, etc.

- New adapter: litellm/llms/you_com/search/transformation.py
  - POSTs to https://ydc-index.io/v1/search
  - Auth: X-API-Key from YOUCOM_API_KEY (or explicit api_key)
  - Maps Perplexity unified spec: max_results -> count,
    search_domain_filter -> include_domains, country -> country
  - Flattens results.web + results.news into a single SearchResult list;
    snippet prefers snippets[0], falls back to description; page_age -> date
- Registry: SearchProviders.YOU_COM in litellm/types/utils.py and wired
  into ProviderConfigManager.get_provider_search_config()
- Pricing entry: model_prices_and_context_window.json (placeholder $0.0;
  happy to adjust to maintainers' preferred public number)
- Docs: example router config snippet and example proxy yaml updated
- Tests: tests/search_tests/test_you_com_search.py - 5 mocked tests
  (payload shape, domain filter mapping, snippet fallback, news flattening,
  missing-api-key error)

Refs upstream expansion signal: BerriAI#15942
@CLAassistant

CLAassistant commented May 20, 2026

Copy link
Copy Markdown

CLA assistant check
All committers have signed the CLA.

@greptile-apps

greptile-apps Bot commented May 20, 2026

Copy link
Copy Markdown
Contributor

Greptile Summary

This PR registers You.com as a new search_provider in the search_tools registry, following the same adapter pattern as Tavily, Exa, Perplexity, and the other existing providers.

  • New adapter (litellm/llms/you_com/search/transformation.py): POSTs to https://ydc-index.io/v1/search, authenticates via X-API-Key, maps the unified Perplexity spec params (max_resultscount, search_domain_filterinclude_domains), and flattens results.web + results.news into a single SearchResult list.
  • Registry wiring: SearchProviders.YOU_COM added to the enum in litellm/types/utils.py and mapped to YouComSearchConfig in ProviderConfigManager.get_provider_search_config() in litellm/utils.py.
  • Pricing placeholder: you_com/search entry added to model_prices_and_context_window.json at $0.0 per query.

Confidence Score: 4/5

Safe to merge; the changes are additive and self-contained, touching only the new provider module and its registry hookup.

The core wiring and response normalisation are correct and mirror existing providers closely. Two small issues in the adapter itself are worth addressing before shipping: the URL normalisation logic can double-append /v1/search if a custom base URL has a trailing slash, and the country parameter is not lowercased while every other provider that supports it does lowercase it. Tests also mutate the process environment without cleanup, which could silently affect unrelated test runs.

litellm/llms/you_com/search/transformation.py (URL construction and country normalisation) and tests/search_tests/test_you_com_search.py (env var cleanup).

Important Files Changed

Filename Overview
litellm/llms/you_com/search/transformation.py New You.com search adapter: transforms queries, maps params, normalises responses. Minor URL-path construction edge case and country case inconsistency.
litellm/types/utils.py Adds YOU_COM = "you_com" to the SearchProviders enum; straightforward and consistent with other entries.
litellm/utils.py Wires YouComSearchConfig into ProviderConfigManager; mirrors the pattern used by all other search providers.
tests/search_tests/test_you_com_search.py Five mock-based tests covering core paths. Sets YOUCOM_API_KEY directly in os.environ without fixture-based cleanup, which can pollute subsequent tests.
model_prices_and_context_window.json Adds you_com/search entry at 0.0 input_cost_per_query; placeholder pricing, correctly structured.
litellm/integrations/websearch_interception/ARCHITECTURE.md Adds you_com example to the provider snippet; documentation change that should be in the litellm-docs repo per project policy.
litellm/proxy/example_config_yaml/websearch_interception_config.yaml Adds commented-out you_com example; harmless documentation comment.

Reviews (1): Last reviewed commit: "feat(search): add you_com as a search pr..." | Re-trigger Greptile

Comment on lines +76 to +77
if not api_base.endswith("/v1/search"):
api_base = f"{api_base.rstrip('/')}/v1/search"

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 The trailing-slash guard and the path-append are applied in the wrong order. endswith("/v1/search") is evaluated on the raw string, so a custom base ending in /v1/search/ fails the check, then rstrip('/') removes only the slash, and the result is …/v1/search/v1/search. The base should be normalised before the check.

Suggested change
if not api_base.endswith("/v1/search"):
api_base = f"{api_base.rstrip('/')}/v1/search"
api_base = api_base.rstrip("/")
if not api_base.endswith("/v1/search"):
api_base = f"{api_base}/v1/search"

Comment on lines +110 to +111
if "country" in optional_params:
request_data["country"] = optional_params["country"]

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 TavilySearchConfig.transform_search_request normalises the country value with .lower() before sending it upstream (see litellm/llms/tavily/search/transformation.py:141). The You.com adapter passes the value through as-is, so a caller using country="US" (uppercase, which is the idiomatic form in the unified spec) may receive different behaviour depending on how You.com's API validates it. Consider applying .lower() here for consistency or documenting the accepted case.

Suggested change
if "country" in optional_params:
request_data["country"] = optional_params["country"]
if "country" in optional_params:
request_data["country"] = optional_params["country"].lower()

Comment on lines +23 to +24
Validate the You.com search request payload structure without real API calls.
"""

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Environment variable leaks between tests

Each async test opens by writing os.environ["YOUCOM_API_KEY"] = "test-api-key" directly and never removes it; test_you_com_search_raises_without_api_key then calls os.environ.pop("YOUCOM_API_KEY", None) to simulate a missing key, but that mutation is permanent for the rest of the session. If test ordering changes or tests are parallelised, this creates false positives or false negatives. Prefer monkeypatch.setenv / monkeypatch.delenv (or a pytest.fixture with yield) so env state is restored automatically after each test.

Comment on lines 244 to 250
- search_tool_name: "my-tavily-tool"
litellm_params:
search_provider: "tavily"
- search_tool_name: "my-you-com-tool"
litellm_params:
search_provider: "you_com"
```

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Documentation changes should live in the litellm-docs repo

Per project policy, documentation additions belong in the external docs repo rather than here. This snippet addition and the corresponding change to litellm/proxy/example_config_yaml/websearch_interception_config.yaml appear to be documentation-only content that should be tracked separately.

Rule Used: Prevent documentation from being added - needs to ... (source)

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!

@codecov

codecov Bot commented May 20, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 95.71429% with 3 lines in your changes missing coverage. Please review.

Files with missing lines Patch % Lines
litellm/llms/you_com/search/transformation.py 95.45% 3 Missing ⚠️

📢 Thoughts on this report? Let us know!

…o test

Addresses Greptile inline review comments on BerriAI#28370:

- get_complete_url: strip trailing slashes from api_base *before* the
  endswith("/v1/search") check, so a custom base like ".../v1/search/"
  doesn't become ".../v1/search/v1/search".
- transform_search_request: .lower() country before sending, matching
  Tavily's convention so callers using the unified spec form ("US") get
  consistent behavior across providers.
- Tests: replace direct os.environ writes with an autouse monkeypatch
  fixture so YOUCOM_API_KEY is set per-test and removed afterwards.
  The missing-key test now uses monkeypatch.delenv. New test asserts the
  trailing-slash normalization above.

Reverts the ARCHITECTURE.md / example yaml edits per the reviewer note
that documentation changes belong in the litellm-docs repo.
@brainsparker

Copy link
Copy Markdown
Contributor Author

Thanks for the review! Pushed fixups in 903e44e:

  • get_complete_url — strip trailing slashes from api_base before the endswith("/v1/search") check, so a custom base like .../v1/search/ doesn't become .../v1/search/v1/search. New test test_you_com_search_complete_url_handles_trailing_slash covers this.
  • Country lowercasetransform_search_request now calls .lower() on country before sending, matching Tavily's behavior. Test updated to assert "us" instead of "US".
  • Test env-var leak — replaced direct os.environ writes with an autouse monkeypatch fixture (scoped per-test). The missing-key test uses monkeypatch.delenv instead of a permanent pop.
  • Docs — reverted the ARCHITECTURE.md and example yaml edits per the note that documentation changes belong in the litellm-docs repo. Happy to send a follow-up PR there if you point me at the repo.

All 6 tests pass locally.

You.com offers an IP-throttled keyless endpoint that returns the same
response shape as the keyed one (~100 queries/day, no signup). This is a
significant onboarding lever - mirrors the keyless DuckDuckGo/SearXNG
providers already in the search_tools registry.

Behavior:
- YOUCOM_API_KEY set        -> keyed:  POST https://ydc-index.io/v1/search
                                       (X-API-Key header)
- no key                    -> free:   POST https://api.you.com/v1/agents/search
                                       (no auth)
- YOUCOM_API_BASE override  -> honored as-is

Tests:
- New: test_you_com_search_keyless_free_tier - asserts URL + absence of
  X-API-Key when no key is configured.
- New: test_you_com_search_validate_environment_keyless - asserts the
  config no longer raises when the key is absent.
- Removed: test_you_com_search_raises_without_api_key (the precondition
  no longer holds).
- Existing payload/domain-filter/etc tests still cover keyed mode via
  the autouse YOUCOM_API_KEY fixture.

Verified both endpoints accept POST + return identical JSON shape:
  results.web[] / results.news[] with title, url, snippets, description,
  page_age.
@brainsparker

Copy link
Copy Markdown
Contributor Author

Added keyless free-tier support in 6eb1912.

If YOUCOM_API_KEY is not set, the adapter now targets api.you.com/v1/agents/search (IP-throttled ~100 queries/day, no signup) instead of erroring. Setting the key upgrades to the keyed endpoint with higher rate limits. Mirrors the existing keyless-default pattern in duckduckgo / searxng.

Mode Endpoint Auth
YOUCOM_API_KEY set POST https://ydc-index.io/v1/search X-API-Key
no key POST https://api.you.com/v1/agents/search none

Both endpoints return the same results.web[] / results.news[] JSON shape (verified against live API). Two new tests (test_you_com_search_keyless_free_tier, test_you_com_search_validate_environment_keyless) cover the keyless path. 7/7 tests pass locally.

PR description updated to reflect the dual-tier behavior.

Adding `litellm/llms/you_com/` requires a corresponding entry in
provider_endpoints_support.json or the
code-quality/check_provider_folders_documented CI check fails.

Follows the compact tavily/serper pattern - endpoints: { search: true }.
Local run of the check now reports "All 114 provider folders are documented".
@krrish-berri-2

Copy link
Copy Markdown
Contributor

Merge Confidence: 3/5 ❌ BLOCKED
⚠️ 1 check failing: codecov/patch
1 PR-related CI failure need fixes first.

Score docked for: 1 PR-related CI failure (codecov/patch).

Drill-down

PR-related failures
• codecov/patch — The new adapter file litellm/llms/you_com/search/transformation.py is in this diff and has 0% patch coverage (65 lines missing); codecov/patch fails because the added production code is not exercised by the CI test suite even though unit tests exist locally.

@krrish-berri-2

Copy link
Copy Markdown
Contributor

please include a video of this working as expected

@krrish-berri-2

Copy link
Copy Markdown
Contributor

you should also file an adjacent pr on litellm-docs so people know this exists

The litellm CI workflows scope unit tests to `tests/test_litellm/...`
(see test-unit-llm-providers.yml: `tests/test_litellm/llms` path), so
tests living under `tests/search_tests/` are never run in CI - which is
why codecov reports 0% patch coverage for the new adapter even though
the unit tests exist and pass locally.

Move test_you_com_search.py into `tests/test_litellm/llms/you_com/` so
the test-unit-llm-providers job picks it up. 7/7 tests still pass at
the new location.

(Sibling search-only providers - tavily, exa_ai, brave, etc. - still
live only in `tests/search_tests/` and would benefit from the same
move, but that is out of scope for this PR.)
@brainsparker

Copy link
Copy Markdown
Contributor Author

Thanks @krrish-berri-2 — addressing your feedback:

1. codecov/patch blocker — root cause was that tests/search_tests/ isn't included in any CI workflow path (all test-unit-* jobs scope to tests/test_litellm/... or tests/proxy_*), so the unit tests existed but CI never ran them. Moved the file to tests/test_litellm/llms/you_com/test_you_com_search.py in 65ac846 so the test-unit-llm-providers job picks it up. 7/7 tests still pass at the new location.

2. Adjacent docs PR — filed at BerriAI/litellm-docs#182 (new docs/search/you_com.md page, sidebar entry, supported-providers table row).

3. Demo video — recording in progress, will attach to this PR shortly.

FYI on (1): every other search-only provider in the registry (tavily, exa_ai, brave, serper, linkup, firecrawl, etc.) currently has the same hidden-from-CI issue — their tests also live only in tests/search_tests/. Happy to send a follow-up PR moving those too if useful, but kept the scope tight here.

The keyless free-tier endpoint (api.you.com/v1/agents/search) advertises
Content-Encoding: gzip but returns a body that httpx's decoder rejects
with `zlib.error: Error -3 while decompressing data: incorrect header
check`, surfacing as litellm.APIConnectionError in user code. curl works
because it doesn't request compression by default.

Pin Accept-Encoding: identity in validate_environment so the upstream
server skips compression entirely. Harmless on the keyed endpoint
(ydc-index.io/v1/search) which negotiates content-encoding correctly.

The header uses setdefault so a caller-supplied Accept-Encoding still
takes precedence. (Server-side bug has been flagged to the You.com team
separately - once fixed there, this workaround can be removed.)

New unit test: test_you_com_search_pins_identity_accept_encoding.
@brainsparker

Copy link
Copy Markdown
Contributor Author

Demo recordings, as promised:

Keyless (no YOUCOM_API_KEY set, routes to api.you.com/v1/agents/search):
asciicast

Keyed (YOUCOM_API_KEY set, routes to ydc-index.io/v1/search):
asciicast

Both hit POST /v1/search/{search_tool_name} with search_provider: you_com configured — exercises YouComSearchConfig via ProviderConfigManager. The endpoint divergence is visible in the --detailed_debug log lines at the end of each cast.

@krrish-berri-2

Copy link
Copy Markdown
Contributor

@Sameerlite can you take over review please

@Sameerlite
Sameerlite changed the base branch from litellm_internal_staging to litellm_oss_staging_050626 June 5, 2026 10:57

@Sameerlite Sameerlite left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thanks!

@Sameerlite
Sameerlite merged commit eab34a3 into BerriAI:litellm_oss_staging_050626 Jun 5, 2026
70 checks passed
Sameerlite added a commit to BerriAI/litellm-docs that referenced this pull request Jun 5, 2026
New page at docs/search/you_com.md documenting both tiers:
- Keyless free tier (default, no signup, IP-throttled)
- Keyed tier via YOUCOM_API_KEY (higher rate limits)

Also updates the search overview index:
- Adds `you_com` to the Supported Providers list
- Adds `you_com` to the search_provider parameter enumeration
- Adds a You.com row to the Supported Providers table

And the sidebar (sidebars.js) so the new page is reachable.

Companion to BerriAI/litellm#28370 (registers `you_com` as a
search_provider in the search_tools registry).

Co-authored-by: Sameer Kankute <sameer@berri.ai>
Sameerlite added a commit that referenced this pull request Jun 5, 2026
* feat(search): add you_com as a search provider

Registers You.com Search API as a first-class `search_provider` in the
`search_tools` registry, alongside Tavily, Exa, Perplexity, etc.

- New adapter: litellm/llms/you_com/search/transformation.py
  - POSTs to https://ydc-index.io/v1/search
  - Auth: X-API-Key from YOUCOM_API_KEY (or explicit api_key)
  - Maps Perplexity unified spec: max_results -> count,
    search_domain_filter -> include_domains, country -> country
  - Flattens results.web + results.news into a single SearchResult list;
    snippet prefers snippets[0], falls back to description; page_age -> date
- Registry: SearchProviders.YOU_COM in litellm/types/utils.py and wired
  into ProviderConfigManager.get_provider_search_config()
- Pricing entry: model_prices_and_context_window.json (placeholder $0.0;
  happy to adjust to maintainers' preferred public number)
- Docs: example router config snippet and example proxy yaml updated
- Tests: tests/search_tests/test_you_com_search.py - 5 mocked tests
  (payload shape, domain filter mapping, snippet fallback, news flattening,
  missing-api-key error)

Refs upstream expansion signal: #15942

* review fixups: normalize api_base, lowercase country, scope env-var to test

Addresses Greptile inline review comments on #28370:

- get_complete_url: strip trailing slashes from api_base *before* the
  endswith("/v1/search") check, so a custom base like ".../v1/search/"
  doesn't become ".../v1/search/v1/search".
- transform_search_request: .lower() country before sending, matching
  Tavily's convention so callers using the unified spec form ("US") get
  consistent behavior across providers.
- Tests: replace direct os.environ writes with an autouse monkeypatch
  fixture so YOUCOM_API_KEY is set per-test and removed afterwards.
  The missing-key test now uses monkeypatch.delenv. New test asserts the
  trailing-slash normalization above.

Reverts the ARCHITECTURE.md / example yaml edits per the reviewer note
that documentation changes belong in the litellm-docs repo.

* support keyless free tier (api.you.com/v1/agents/search) as default

You.com offers an IP-throttled keyless endpoint that returns the same
response shape as the keyed one (~100 queries/day, no signup). This is a
significant onboarding lever - mirrors the keyless DuckDuckGo/SearXNG
providers already in the search_tools registry.

Behavior:
- YOUCOM_API_KEY set        -> keyed:  POST https://ydc-index.io/v1/search
                                       (X-API-Key header)
- no key                    -> free:   POST https://api.you.com/v1/agents/search
                                       (no auth)
- YOUCOM_API_BASE override  -> honored as-is

Tests:
- New: test_you_com_search_keyless_free_tier - asserts URL + absence of
  X-API-Key when no key is configured.
- New: test_you_com_search_validate_environment_keyless - asserts the
  config no longer raises when the key is absent.
- Removed: test_you_com_search_raises_without_api_key (the precondition
  no longer holds).
- Existing payload/domain-filter/etc tests still cover keyed mode via
  the autouse YOUCOM_API_KEY fixture.

Verified both endpoints accept POST + return identical JSON shape:
  results.web[] / results.news[] with title, url, snippets, description,
  page_age.

* register you_com in provider_endpoints_support.json

Adding `litellm/llms/you_com/` requires a corresponding entry in
provider_endpoints_support.json or the
code-quality/check_provider_folders_documented CI check fails.

Follows the compact tavily/serper pattern - endpoints: { search: true }.
Local run of the check now reports "All 114 provider folders are documented".

* move tests under tests/test_litellm/llms/ so CI exercises them

The litellm CI workflows scope unit tests to `tests/test_litellm/...`
(see test-unit-llm-providers.yml: `tests/test_litellm/llms` path), so
tests living under `tests/search_tests/` are never run in CI - which is
why codecov reports 0% patch coverage for the new adapter even though
the unit tests exist and pass locally.

Move test_you_com_search.py into `tests/test_litellm/llms/you_com/` so
the test-unit-llm-providers job picks it up. 7/7 tests still pass at
the new location.

(Sibling search-only providers - tavily, exa_ai, brave, etc. - still
live only in `tests/search_tests/` and would benefit from the same
move, but that is out of scope for this PR.)

* fix(you_com): pin Accept-Encoding: identity to dodge keyless gzip bug

The keyless free-tier endpoint (api.you.com/v1/agents/search) advertises
Content-Encoding: gzip but returns a body that httpx's decoder rejects
with `zlib.error: Error -3 while decompressing data: incorrect header
check`, surfacing as litellm.APIConnectionError in user code. curl works
because it doesn't request compression by default.

Pin Accept-Encoding: identity in validate_environment so the upstream
server skips compression entirely. Harmless on the keyed endpoint
(ydc-index.io/v1/search) which negotiates content-encoding correctly.

The header uses setdefault so a caller-supplied Accept-Encoding still
takes precedence. (Server-side bug has been flagged to the You.com team
separately - once fixed there, this workaround can be removed.)

New unit test: test_you_com_search_pins_identity_accept_encoding.

---------

Co-authored-by: Sameer Kankute <sameer@berri.ai>
@brainsparker
brainsparker deleted the you-com-search-provider branch June 5, 2026 17:49
mateo-berri added a commit that referenced this pull request Jun 5, 2026
* Mark xAI models retiring on 2026-05-15 (#28788)

Per https://docs.x.ai/developers/migration/may-15-retirement, xAI is
retiring the following slugs on 2026-05-15 (auto-redirect to grok-4.3
with various reasoning efforts; callers continuing to use the old slugs
will be billed at grok-4.3 pricing):

  grok-4-1-fast-reasoning{,-latest}      -> grok-4.3 (low effort)
  grok-4-1-fast-non-reasoning{,-latest}  -> grok-4.3 (none)
  grok-4-fast-reasoning                  -> grok-4.3 (low effort)
  grok-4-fast-non-reasoning              -> grok-4.3 (none)
  grok-4-0709                            -> grok-4.3 (low effort)
  grok-code-fast-1{,-0825}               -> grok-build-0.1
  grok-3                                 -> grok-4.3 (none)

Only the direct xai/ slugs are tagged; third-party hosts (azure_ai,
oci, vercel_ai_gateway, perplexity/xai) run their own schedules. The
grok-3 retirement list explicitly names only the base grok-3 slug — the
-mini / -fast / -beta / -latest variants are not listed, so they remain
untouched.

* feat(moonshot): advertise json_schema response support on live models (#29683)

litellm.responses() already routes Moonshot through the responses->chat-completions
bridge, and Moonshot honors response_format json_schema on chat completions. The
cost-map entries left supports_response_schema unset, so discovery layers that gate
on that flag dropped Moonshot from structured-output / responses listings even though
the capability works end to end.

Set supports_response_schema on the nine models currently live on api.moonshot.ai:
kimi-k2.5, kimi-k2.6, the moonshot-v1 8k/32k/128k text and vision-preview variants,
and moonshot-v1-auto. Verified against the live API that each honors json_schema and
that litellm.responses() returns schema-valid structured output through the bridge.

* chore(moonshot): mark models retired from api.moonshot.ai as deprecated (#29685)

Thirteen Moonshot/Kimi models in the cost map no longer resolve on
api.moonshot.ai (all return 404). Stamp each with its deprecation_date from
platform.kimi.ai/docs/models rather than deleting the entries, so historical
cost calculation keeps resolving the names while tooling can surface the
retirement.

Dates: kimi-thinking-preview 2025-11-11; kimi-latest and its 8k/32k/128k context
variants 2026-01-28; the kimi-k2 preview/turbo/thinking series 2026-05-25; the
moonshot-v1 -0430 snapshots use their own 2024-04-30 snapshot date (Moonshot
publishes no discontinuation date for them).

* fix(moonshot): drop temperature for reasoning models (kimi-k2.5/k2.6) (#29687)

Kimi reasoning models reject every temperature except 1; a request with
temperature=0.2 returns "invalid temperature: only 1 is allowed for this model".
litellm only clamped temperature into [0.3, 1], so any value below 1 still 400'd.

Drop the temperature param entirely for reasoning models (gated on
supports_reasoning, the same signal transform_request already uses) so the model
default is used; the non-reasoning moonshot-v1 models keep the existing clamp.

Co-authored-by: Sameer Kankute <sameer@berri.ai>

* feat(mcp): add per-server timeout configuration (#29672)

* feat(mcp): add per-server timeout configuration

* fix(mcp): address timeout field review comments

- use is not None guard instead of or for 0.0 edge case
- copy timeout in both LiteLLM_MCPServerTable constructions (health check path + _build_mcp_server_table)
- add timeout Float? column to all three schema.prisma files
- extend round-trip test to cover _build_mcp_server_table direction
- add test for zero timeout not treated as falsy

* fix(mcp): forward timeout in _build_temporary_mcp_server_record

* fix(mcp): return 504 instead of 500 when per-server timeout fires

* test(mcp): add 504 timeout regression test; fix black formatting

* Add jp. Bedrock cross-region inference profile for claude-opus-4-7 (#28567)

* fix(thinking): handle None thinking param in is_thinking_enabled (#28598)

Squash-merged by litellm-agent from Terrajlz's PR.

* feat(helm): support tpl rendering in podAnnotations (#28609)

Squash-merged by litellm-agent from devauxbr's PR.

* Forward custom_llm_provider through the Responses API bridge (Fixes #28505) (#28575)

* Forward custom_llm_provider through the Responses API bridge (Fixes #28505)

When a Chat Completions request to a GPT-5.4+ model contains both
`tools` and `reasoning_effort`, `completion()` auto-routes through
`responses_api_bridge`. The bridge handler called
`litellm.responses()` / `litellm.aresponses()` without forwarding the
already-resolved `custom_llm_provider`, so the downstream call
re-invoked `get_llm_provider()` with `custom_llm_provider=None` and
stripped a second provider prefix from a `provider/provider/model`
deployment string.

For a deployment configured as `openai/openai/openai/gpt-5.5`,
the bridge flow sent `openai/gpt-5.5` to the upstream API instead of
the correct `openai/openai/gpt-5.5`. Upstream APIs that enforce
model-name allow-lists rejected this as `key_model_access_denied`.

Fix: pass the locally-resolved `custom_llm_provider` into both the
sync `responses()` and async `aresponses()` calls so the downstream
`_resolve_model_provider_for_responses` sees an explicit provider
and skips the second prefix-strip.

New regression test
`tests/test_litellm/completion_extras/test_responses_bridge_provider_propagation.py`
pins both call sites: each must forward `custom_llm_provider`.

* fix(28505): set custom_llm_provider on request_data instead of as duplicate kwarg

Greptile flagged that the previous patch passed custom_llm_provider as an
explicit kwarg to responses()/aresponses() while request_data already
carried it via the spread of sanitized_litellm_params, which would raise
TypeError: got multiple values for keyword argument on every real bridge
call.

Switches to assigning request_data['custom_llm_provider'] before the call
so the resolved provider wins over whatever sanitized_litellm_params spread
in, without duplicating the kwarg.

Updates the regression test to seed request_data with a sentinel
custom_llm_provider so it actually exercises the overwrite path (the
previous test mocked transform_request with a minimal dict and never hit
the conflict).

* chore: trigger shin-agent re-eval on retargeted staging base

* chore: trigger shin-agent re-eval against updated Greptile state

* Add jp. Bedrock cross-region inference profile for claude-opus-4-7

AWS Bedrock documents jp.anthropic.claude-opus-4-7 alongside the
existing us./eu./au./global. profiles for Claude Opus 4.7
(ap-northeast-1 Tokyo / ap-northeast-3 Osaka), but the entry is
missing from model_prices_and_context_window.json. Tokyo-region
users currently get an "unknown model" error when routing through
the JP geo profile.

Adds the entry to both the canonical file and the bundled backup,
mirroring the recent pattern for sonnet-4-6 (#27831). Pricing matches
the other regional profiles (10% premium over base/global).

Regression test pins all six documented profiles (base, global, us, eu,
au, jp) and asserts pricing parity between jp. and au. variants.

Source: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-7.html

---------

Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: Sameer Kankute <sameer@berri.ai>

* feat(soniox): add soniox audio transcription integration (#29508)

* feat(openmeter): add OPENMETER_TRUST_REQUEST_USER to prevent forged attribution (#29650)

The OpenMeter callback resolves the CloudEvent subject from kwargs["user"]
first, then falls back to the key-bound user_api_key_user_id. For
multi-tenant proxy deployments, a client can set `"user": "..."` in the
request body and cause their usage to be attributed to that arbitrary
string — a billing-attribution forgery risk.

Adds OPENMETER_TRUST_REQUEST_USER env var (default "true" for backward
compatibility). When set to "false", the request-supplied `user` field is
ignored and the subject is resolved solely from user_api_key_user_id.

Matches the existing env-var-driven config pattern in this file
(OPENMETER_API_KEY, OPENMETER_API_ENDPOINT, OPENMETER_EVENT_TYPE).

* feat(search): add you_com as a search provider (#28370)

* feat(search): add you_com as a search provider

Registers You.com Search API as a first-class `search_provider` in the
`search_tools` registry, alongside Tavily, Exa, Perplexity, etc.

- New adapter: litellm/llms/you_com/search/transformation.py
  - POSTs to https://ydc-index.io/v1/search
  - Auth: X-API-Key from YOUCOM_API_KEY (or explicit api_key)
  - Maps Perplexity unified spec: max_results -> count,
    search_domain_filter -> include_domains, country -> country
  - Flattens results.web + results.news into a single SearchResult list;
    snippet prefers snippets[0], falls back to description; page_age -> date
- Registry: SearchProviders.YOU_COM in litellm/types/utils.py and wired
  into ProviderConfigManager.get_provider_search_config()
- Pricing entry: model_prices_and_context_window.json (placeholder $0.0;
  happy to adjust to maintainers' preferred public number)
- Docs: example router config snippet and example proxy yaml updated
- Tests: tests/search_tests/test_you_com_search.py - 5 mocked tests
  (payload shape, domain filter mapping, snippet fallback, news flattening,
  missing-api-key error)

Refs upstream expansion signal: #15942

* review fixups: normalize api_base, lowercase country, scope env-var to test

Addresses Greptile inline review comments on #28370:

- get_complete_url: strip trailing slashes from api_base *before* the
  endswith("/v1/search") check, so a custom base like ".../v1/search/"
  doesn't become ".../v1/search/v1/search".
- transform_search_request: .lower() country before sending, matching
  Tavily's convention so callers using the unified spec form ("US") get
  consistent behavior across providers.
- Tests: replace direct os.environ writes with an autouse monkeypatch
  fixture so YOUCOM_API_KEY is set per-test and removed afterwards.
  The missing-key test now uses monkeypatch.delenv. New test asserts the
  trailing-slash normalization above.

Reverts the ARCHITECTURE.md / example yaml edits per the reviewer note
that documentation changes belong in the litellm-docs repo.

* support keyless free tier (api.you.com/v1/agents/search) as default

You.com offers an IP-throttled keyless endpoint that returns the same
response shape as the keyed one (~100 queries/day, no signup). This is a
significant onboarding lever - mirrors the keyless DuckDuckGo/SearXNG
providers already in the search_tools registry.

Behavior:
- YOUCOM_API_KEY set        -> keyed:  POST https://ydc-index.io/v1/search
                                       (X-API-Key header)
- no key                    -> free:   POST https://api.you.com/v1/agents/search
                                       (no auth)
- YOUCOM_API_BASE override  -> honored as-is

Tests:
- New: test_you_com_search_keyless_free_tier - asserts URL + absence of
  X-API-Key when no key is configured.
- New: test_you_com_search_validate_environment_keyless - asserts the
  config no longer raises when the key is absent.
- Removed: test_you_com_search_raises_without_api_key (the precondition
  no longer holds).
- Existing payload/domain-filter/etc tests still cover keyed mode via
  the autouse YOUCOM_API_KEY fixture.

Verified both endpoints accept POST + return identical JSON shape:
  results.web[] / results.news[] with title, url, snippets, description,
  page_age.

* register you_com in provider_endpoints_support.json

Adding `litellm/llms/you_com/` requires a corresponding entry in
provider_endpoints_support.json or the
code-quality/check_provider_folders_documented CI check fails.

Follows the compact tavily/serper pattern - endpoints: { search: true }.
Local run of the check now reports "All 114 provider folders are documented".

* move tests under tests/test_litellm/llms/ so CI exercises them

The litellm CI workflows scope unit tests to `tests/test_litellm/...`
(see test-unit-llm-providers.yml: `tests/test_litellm/llms` path), so
tests living under `tests/search_tests/` are never run in CI - which is
why codecov reports 0% patch coverage for the new adapter even though
the unit tests exist and pass locally.

Move test_you_com_search.py into `tests/test_litellm/llms/you_com/` so
the test-unit-llm-providers job picks it up. 7/7 tests still pass at
the new location.

(Sibling search-only providers - tavily, exa_ai, brave, etc. - still
live only in `tests/search_tests/` and would benefit from the same
move, but that is out of scope for this PR.)

* fix(you_com): pin Accept-Encoding: identity to dodge keyless gzip bug

The keyless free-tier endpoint (api.you.com/v1/agents/search) advertises
Content-Encoding: gzip but returns a body that httpx's decoder rejects
with `zlib.error: Error -3 while decompressing data: incorrect header
check`, surfacing as litellm.APIConnectionError in user code. curl works
because it doesn't request compression by default.

Pin Accept-Encoding: identity in validate_environment so the upstream
server skips compression entirely. Harmless on the keyed endpoint
(ydc-index.io/v1/search) which negotiates content-encoding correctly.

The header uses setdefault so a caller-supplied Accept-Encoding still
takes precedence. (Server-side bug has been flagged to the You.com team
separately - once fixed there, this workaround can be removed.)

New unit test: test_you_com_search_pins_identity_accept_encoding.

---------

Co-authored-by: Sameer Kankute <sameer@berri.ai>

* docs: fix README typo (#29419)

Correct clear spelling mistakes in documentation without changing behavior.

Confidence: high
Scope-risk: narrow
Tested: git diff --check; uvx codespell on changed files
Not-tested: Full docs build not run; text-only changes

* Fix(langfuse): pass httpx_client to Langfuse in langfuse_prompt_management to respect SSL_VERIFY (#29480)

* fix(langfuse): pass ssl_verify to Langfuse httpx client

* fix_langfuse_

* add unit tests

* addressed comments

---------

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

* feat(models): add minimax/MiniMax-M3 to model cost map (#29412)

Add MiniMax's new flagship MiniMax-M3 to the native minimax provider:
512K context, 128K max output, native multimodal (supports_vision),
reasoning, prompt caching. Pricing (USD/M tokens): input 0.6 / output
2.4 / cache read 0.12. M3 has no active prompt-cache-write tier, so
cache_creation_input_token_cost is omitted.

Updated both the root model_prices_and_context_window.json (remote
source) and the bundled litellm/model_prices_and_context_window_backup.json
(local fallback), keeping them in sync.

* fix(logging): handle ResponseCompletedEvent in anthropic_messages streaming spend log (#29394)

* fix(logging): handle ResponseCompletedEvent in anthropic_messages streaming spend log

* fix(logging): extend terminal event handling to ResponseIncompleteEvent and ResponseFailedEvent; fix return type annotation

* feat(provider): Add Neosantara provider as OpenAI Compatible (#29646)

* Add Neosantara provider

* Register Neosantara provider enum

* Address Neosantara provider review feedback

* Add Neosantara packaged endpoint support

---------

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

* fix: address greptile and veria review feedback

- langfuse: guard httpx_client injection behind version check (>= 2.7.3)
- soniox: propagate audio_transcription_duration in _hidden_params for spend tracking
- soniox: give SONIOX_API_BASE env var priority over caller-supplied api_base
- mcp: replace CancelledError catch with asyncio.wait_for + TimeoutError

* chore(mcp): add migration for per-server timeout column

* fix(test): add tool_use_system_prompt_tokens to model prices schema validator

* fix: mcp timeout test uses real asyncio.wait_for timeout; you_com get_complete_url respects resolved api_key

* fix: forward resolved api_key into you_com endpoint selection and apply timeout to soniox polling GETs

The search flow resolves api_key in validate_environment but never passed it
into get_complete_url, so a programmatic api_key (with no YOUCOM_API_KEY in the
env) set the X-API-Key header yet still selected the keyless free-tier endpoint.
Forward api_key through both the search entrypoint and the http handler so the
keyed endpoint is chosen.

HTTPHandler.get/AsyncHTTPHandler.get had no timeout parameter, so the Soniox
poll and transcript-fetch GETs silently used the client global default instead
of the caller timeout. Add a per-request timeout to get() and forward the
configured timeout from the Soniox handler.

* fix(soniox): price stt-async-v4 per second so transcriptions are billed

The handler stores audio_transcription_duration in _hidden_params, but the
model carried only token cost fields and the response has no token usage, so
the transcription cost path fell through to cost_per_second and returned $0.
An authenticated caller could transcribe Soniox audio without decrementing
their budget. Switch the entry to output_cost_per_second at Soniox's published
$0.10/hour async rate so the stored duration produces a real charge.

* fix(langfuse): use a dedicated httpx client for the SDK injection

The httpx_client handed to the Langfuse SDK came from _get_httpx_client(),
which returns LiteLLM's globally cached HTTPHandler. If Langfuse closed that
client on teardown it would invalidate the shared client used by every other
LiteLLM HTTP call. Build a dedicated httpx.Client instead, still resolving SSL
verification and client certificate from LiteLLM's configuration.

* fix(soniox): prefer caller-supplied api_base over SONIOX_API_BASE env var

* fix(cohere): support max_completion_tokens on cohere v2 chat (default route) (#29779)

* fix(cohere): support max_completion_tokens on cohere v2 chat

The default cohere_chat route resolves to CohereV2ChatConfig, which did not
list or map max_completion_tokens, so get_optional_params raised
UnsupportedParamsError for the standard OpenAI parameter (the modern
replacement for the deprecated max_tokens). The v1 config already maps it to
cohere's max_tokens; mirror that in v2 and add v2 regression tests.

* fix(cohere): make max_completion_tokens take precedence over max_tokens on v2

When both max_tokens and max_completion_tokens are supplied, prefer
max_completion_tokens explicitly rather than relying on dict iteration order,
and cover both orderings with a regression test.

---------

Co-authored-by: Daniel Yudelevich <4537920+yudelevi@users.noreply.github.com>
Co-authored-by: hectorc98 <hector.chamorroalvarez@adyen.com>
Co-authored-by: Filippo Menghi <113345637+Cyberfilo@users.noreply.github.com>
Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: Dan Lemon <dan@danlemon.com>
Co-authored-by: Saswat <saswatds@users.noreply.github.com>
Co-authored-by: Brian Sparker <brainsparker@users.noreply.github.com>
Co-authored-by: Zhao73 <156770117+Zhao73@users.noreply.github.com>
Co-authored-by: Urain Ahmad Shah <60431964+urainshah@users.noreply.github.com>
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Co-authored-by: yuneng-jiang <yuneng@berri.ai>
Co-authored-by: kape <168134658+kapelame@users.noreply.github.com>
Co-authored-by: danisalvaa <159898202+danisalvaa@users.noreply.github.com>
Co-authored-by: Just R <remixingmagelang@gmail.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Co-authored-by: abhay23-AI <abhaytrivedi22@gmail.com>
fzowl pushed a commit to fzowl/litellm that referenced this pull request Jun 24, 2026
* Mark xAI models retiring on 2026-05-15 (BerriAI#28788)

Per https://docs.x.ai/developers/migration/may-15-retirement, xAI is
retiring the following slugs on 2026-05-15 (auto-redirect to grok-4.3
with various reasoning efforts; callers continuing to use the old slugs
will be billed at grok-4.3 pricing):

  grok-4-1-fast-reasoning{,-latest}      -> grok-4.3 (low effort)
  grok-4-1-fast-non-reasoning{,-latest}  -> grok-4.3 (none)
  grok-4-fast-reasoning                  -> grok-4.3 (low effort)
  grok-4-fast-non-reasoning              -> grok-4.3 (none)
  grok-4-0709                            -> grok-4.3 (low effort)
  grok-code-fast-1{,-0825}               -> grok-build-0.1
  grok-3                                 -> grok-4.3 (none)

Only the direct xai/ slugs are tagged; third-party hosts (azure_ai,
oci, vercel_ai_gateway, perplexity/xai) run their own schedules. The
grok-3 retirement list explicitly names only the base grok-3 slug — the
-mini / -fast / -beta / -latest variants are not listed, so they remain
untouched.

* feat(moonshot): advertise json_schema response support on live models (BerriAI#29683)

litellm.responses() already routes Moonshot through the responses->chat-completions
bridge, and Moonshot honors response_format json_schema on chat completions. The
cost-map entries left supports_response_schema unset, so discovery layers that gate
on that flag dropped Moonshot from structured-output / responses listings even though
the capability works end to end.

Set supports_response_schema on the nine models currently live on api.moonshot.ai:
kimi-k2.5, kimi-k2.6, the moonshot-v1 8k/32k/128k text and vision-preview variants,
and moonshot-v1-auto. Verified against the live API that each honors json_schema and
that litellm.responses() returns schema-valid structured output through the bridge.

* chore(moonshot): mark models retired from api.moonshot.ai as deprecated (BerriAI#29685)

Thirteen Moonshot/Kimi models in the cost map no longer resolve on
api.moonshot.ai (all return 404). Stamp each with its deprecation_date from
platform.kimi.ai/docs/models rather than deleting the entries, so historical
cost calculation keeps resolving the names while tooling can surface the
retirement.

Dates: kimi-thinking-preview 2025-11-11; kimi-latest and its 8k/32k/128k context
variants 2026-01-28; the kimi-k2 preview/turbo/thinking series 2026-05-25; the
moonshot-v1 -0430 snapshots use their own 2024-04-30 snapshot date (Moonshot
publishes no discontinuation date for them).

* fix(moonshot): drop temperature for reasoning models (kimi-k2.5/k2.6) (BerriAI#29687)

Kimi reasoning models reject every temperature except 1; a request with
temperature=0.2 returns "invalid temperature: only 1 is allowed for this model".
litellm only clamped temperature into [0.3, 1], so any value below 1 still 400'd.

Drop the temperature param entirely for reasoning models (gated on
supports_reasoning, the same signal transform_request already uses) so the model
default is used; the non-reasoning moonshot-v1 models keep the existing clamp.

Co-authored-by: Sameer Kankute <sameer@berri.ai>

* feat(mcp): add per-server timeout configuration (BerriAI#29672)

* feat(mcp): add per-server timeout configuration

* fix(mcp): address timeout field review comments

- use is not None guard instead of or for 0.0 edge case
- copy timeout in both LiteLLM_MCPServerTable constructions (health check path + _build_mcp_server_table)
- add timeout Float? column to all three schema.prisma files
- extend round-trip test to cover _build_mcp_server_table direction
- add test for zero timeout not treated as falsy

* fix(mcp): forward timeout in _build_temporary_mcp_server_record

* fix(mcp): return 504 instead of 500 when per-server timeout fires

* test(mcp): add 504 timeout regression test; fix black formatting

* Add jp. Bedrock cross-region inference profile for claude-opus-4-7 (BerriAI#28567)

* fix(thinking): handle None thinking param in is_thinking_enabled (BerriAI#28598)

Squash-merged by litellm-agent from Terrajlz's PR.

* feat(helm): support tpl rendering in podAnnotations (BerriAI#28609)

Squash-merged by litellm-agent from devauxbr's PR.

* Forward custom_llm_provider through the Responses API bridge (Fixes BerriAI#28505) (BerriAI#28575)

* Forward custom_llm_provider through the Responses API bridge (Fixes BerriAI#28505)

When a Chat Completions request to a GPT-5.4+ model contains both
`tools` and `reasoning_effort`, `completion()` auto-routes through
`responses_api_bridge`. The bridge handler called
`litellm.responses()` / `litellm.aresponses()` without forwarding the
already-resolved `custom_llm_provider`, so the downstream call
re-invoked `get_llm_provider()` with `custom_llm_provider=None` and
stripped a second provider prefix from a `provider/provider/model`
deployment string.

For a deployment configured as `openai/openai/openai/gpt-5.5`,
the bridge flow sent `openai/gpt-5.5` to the upstream API instead of
the correct `openai/openai/gpt-5.5`. Upstream APIs that enforce
model-name allow-lists rejected this as `key_model_access_denied`.

Fix: pass the locally-resolved `custom_llm_provider` into both the
sync `responses()` and async `aresponses()` calls so the downstream
`_resolve_model_provider_for_responses` sees an explicit provider
and skips the second prefix-strip.

New regression test
`tests/test_litellm/completion_extras/test_responses_bridge_provider_propagation.py`
pins both call sites: each must forward `custom_llm_provider`.

* fix(28505): set custom_llm_provider on request_data instead of as duplicate kwarg

Greptile flagged that the previous patch passed custom_llm_provider as an
explicit kwarg to responses()/aresponses() while request_data already
carried it via the spread of sanitized_litellm_params, which would raise
TypeError: got multiple values for keyword argument on every real bridge
call.

Switches to assigning request_data['custom_llm_provider'] before the call
so the resolved provider wins over whatever sanitized_litellm_params spread
in, without duplicating the kwarg.

Updates the regression test to seed request_data with a sentinel
custom_llm_provider so it actually exercises the overwrite path (the
previous test mocked transform_request with a minimal dict and never hit
the conflict).

* chore: trigger shin-agent re-eval on retargeted staging base

* chore: trigger shin-agent re-eval against updated Greptile state

* Add jp. Bedrock cross-region inference profile for claude-opus-4-7

AWS Bedrock documents jp.anthropic.claude-opus-4-7 alongside the
existing us./eu./au./global. profiles for Claude Opus 4.7
(ap-northeast-1 Tokyo / ap-northeast-3 Osaka), but the entry is
missing from model_prices_and_context_window.json. Tokyo-region
users currently get an "unknown model" error when routing through
the JP geo profile.

Adds the entry to both the canonical file and the bundled backup,
mirroring the recent pattern for sonnet-4-6 (BerriAI#27831). Pricing matches
the other regional profiles (10% premium over base/global).

Regression test pins all six documented profiles (base, global, us, eu,
au, jp) and asserts pricing parity between jp. and au. variants.

Source: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-7.html

---------

Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: Sameer Kankute <sameer@berri.ai>

* feat(soniox): add soniox audio transcription integration (BerriAI#29508)

* feat(openmeter): add OPENMETER_TRUST_REQUEST_USER to prevent forged attribution (BerriAI#29650)

The OpenMeter callback resolves the CloudEvent subject from kwargs["user"]
first, then falls back to the key-bound user_api_key_user_id. For
multi-tenant proxy deployments, a client can set `"user": "..."` in the
request body and cause their usage to be attributed to that arbitrary
string — a billing-attribution forgery risk.

Adds OPENMETER_TRUST_REQUEST_USER env var (default "true" for backward
compatibility). When set to "false", the request-supplied `user` field is
ignored and the subject is resolved solely from user_api_key_user_id.

Matches the existing env-var-driven config pattern in this file
(OPENMETER_API_KEY, OPENMETER_API_ENDPOINT, OPENMETER_EVENT_TYPE).

* feat(search): add you_com as a search provider (BerriAI#28370)

* feat(search): add you_com as a search provider

Registers You.com Search API as a first-class `search_provider` in the
`search_tools` registry, alongside Tavily, Exa, Perplexity, etc.

- New adapter: litellm/llms/you_com/search/transformation.py
  - POSTs to https://ydc-index.io/v1/search
  - Auth: X-API-Key from YOUCOM_API_KEY (or explicit api_key)
  - Maps Perplexity unified spec: max_results -> count,
    search_domain_filter -> include_domains, country -> country
  - Flattens results.web + results.news into a single SearchResult list;
    snippet prefers snippets[0], falls back to description; page_age -> date
- Registry: SearchProviders.YOU_COM in litellm/types/utils.py and wired
  into ProviderConfigManager.get_provider_search_config()
- Pricing entry: model_prices_and_context_window.json (placeholder $0.0;
  happy to adjust to maintainers' preferred public number)
- Docs: example router config snippet and example proxy yaml updated
- Tests: tests/search_tests/test_you_com_search.py - 5 mocked tests
  (payload shape, domain filter mapping, snippet fallback, news flattening,
  missing-api-key error)

Refs upstream expansion signal: BerriAI#15942

* review fixups: normalize api_base, lowercase country, scope env-var to test

Addresses Greptile inline review comments on BerriAI#28370:

- get_complete_url: strip trailing slashes from api_base *before* the
  endswith("/v1/search") check, so a custom base like ".../v1/search/"
  doesn't become ".../v1/search/v1/search".
- transform_search_request: .lower() country before sending, matching
  Tavily's convention so callers using the unified spec form ("US") get
  consistent behavior across providers.
- Tests: replace direct os.environ writes with an autouse monkeypatch
  fixture so YOUCOM_API_KEY is set per-test and removed afterwards.
  The missing-key test now uses monkeypatch.delenv. New test asserts the
  trailing-slash normalization above.

Reverts the ARCHITECTURE.md / example yaml edits per the reviewer note
that documentation changes belong in the litellm-docs repo.

* support keyless free tier (api.you.com/v1/agents/search) as default

You.com offers an IP-throttled keyless endpoint that returns the same
response shape as the keyed one (~100 queries/day, no signup). This is a
significant onboarding lever - mirrors the keyless DuckDuckGo/SearXNG
providers already in the search_tools registry.

Behavior:
- YOUCOM_API_KEY set        -> keyed:  POST https://ydc-index.io/v1/search
                                       (X-API-Key header)
- no key                    -> free:   POST https://api.you.com/v1/agents/search
                                       (no auth)
- YOUCOM_API_BASE override  -> honored as-is

Tests:
- New: test_you_com_search_keyless_free_tier - asserts URL + absence of
  X-API-Key when no key is configured.
- New: test_you_com_search_validate_environment_keyless - asserts the
  config no longer raises when the key is absent.
- Removed: test_you_com_search_raises_without_api_key (the precondition
  no longer holds).
- Existing payload/domain-filter/etc tests still cover keyed mode via
  the autouse YOUCOM_API_KEY fixture.

Verified both endpoints accept POST + return identical JSON shape:
  results.web[] / results.news[] with title, url, snippets, description,
  page_age.

* register you_com in provider_endpoints_support.json

Adding `litellm/llms/you_com/` requires a corresponding entry in
provider_endpoints_support.json or the
code-quality/check_provider_folders_documented CI check fails.

Follows the compact tavily/serper pattern - endpoints: { search: true }.
Local run of the check now reports "All 114 provider folders are documented".

* move tests under tests/test_litellm/llms/ so CI exercises them

The litellm CI workflows scope unit tests to `tests/test_litellm/...`
(see test-unit-llm-providers.yml: `tests/test_litellm/llms` path), so
tests living under `tests/search_tests/` are never run in CI - which is
why codecov reports 0% patch coverage for the new adapter even though
the unit tests exist and pass locally.

Move test_you_com_search.py into `tests/test_litellm/llms/you_com/` so
the test-unit-llm-providers job picks it up. 7/7 tests still pass at
the new location.

(Sibling search-only providers - tavily, exa_ai, brave, etc. - still
live only in `tests/search_tests/` and would benefit from the same
move, but that is out of scope for this PR.)

* fix(you_com): pin Accept-Encoding: identity to dodge keyless gzip bug

The keyless free-tier endpoint (api.you.com/v1/agents/search) advertises
Content-Encoding: gzip but returns a body that httpx's decoder rejects
with `zlib.error: Error -3 while decompressing data: incorrect header
check`, surfacing as litellm.APIConnectionError in user code. curl works
because it doesn't request compression by default.

Pin Accept-Encoding: identity in validate_environment so the upstream
server skips compression entirely. Harmless on the keyed endpoint
(ydc-index.io/v1/search) which negotiates content-encoding correctly.

The header uses setdefault so a caller-supplied Accept-Encoding still
takes precedence. (Server-side bug has been flagged to the You.com team
separately - once fixed there, this workaround can be removed.)

New unit test: test_you_com_search_pins_identity_accept_encoding.

---------

Co-authored-by: Sameer Kankute <sameer@berri.ai>

* docs: fix README typo (BerriAI#29419)

Correct clear spelling mistakes in documentation without changing behavior.

Confidence: high
Scope-risk: narrow
Tested: git diff --check; uvx codespell on changed files
Not-tested: Full docs build not run; text-only changes

* Fix(langfuse): pass httpx_client to Langfuse in langfuse_prompt_management to respect SSL_VERIFY (BerriAI#29480)

* fix(langfuse): pass ssl_verify to Langfuse httpx client

* fix_langfuse_

* add unit tests

* addressed comments

---------

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

* feat(models): add minimax/MiniMax-M3 to model cost map (BerriAI#29412)

Add MiniMax's new flagship MiniMax-M3 to the native minimax provider:
512K context, 128K max output, native multimodal (supports_vision),
reasoning, prompt caching. Pricing (USD/M tokens): input 0.6 / output
2.4 / cache read 0.12. M3 has no active prompt-cache-write tier, so
cache_creation_input_token_cost is omitted.

Updated both the root model_prices_and_context_window.json (remote
source) and the bundled litellm/model_prices_and_context_window_backup.json
(local fallback), keeping them in sync.

* fix(logging): handle ResponseCompletedEvent in anthropic_messages streaming spend log (BerriAI#29394)

* fix(logging): handle ResponseCompletedEvent in anthropic_messages streaming spend log

* fix(logging): extend terminal event handling to ResponseIncompleteEvent and ResponseFailedEvent; fix return type annotation

* feat(provider): Add Neosantara provider as OpenAI Compatible (BerriAI#29646)

* Add Neosantara provider

* Register Neosantara provider enum

* Address Neosantara provider review feedback

* Add Neosantara packaged endpoint support

---------

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

* fix: address greptile and veria review feedback

- langfuse: guard httpx_client injection behind version check (>= 2.7.3)
- soniox: propagate audio_transcription_duration in _hidden_params for spend tracking
- soniox: give SONIOX_API_BASE env var priority over caller-supplied api_base
- mcp: replace CancelledError catch with asyncio.wait_for + TimeoutError

* chore(mcp): add migration for per-server timeout column

* fix(test): add tool_use_system_prompt_tokens to model prices schema validator

* fix: mcp timeout test uses real asyncio.wait_for timeout; you_com get_complete_url respects resolved api_key

* fix: forward resolved api_key into you_com endpoint selection and apply timeout to soniox polling GETs

The search flow resolves api_key in validate_environment but never passed it
into get_complete_url, so a programmatic api_key (with no YOUCOM_API_KEY in the
env) set the X-API-Key header yet still selected the keyless free-tier endpoint.
Forward api_key through both the search entrypoint and the http handler so the
keyed endpoint is chosen.

HTTPHandler.get/AsyncHTTPHandler.get had no timeout parameter, so the Soniox
poll and transcript-fetch GETs silently used the client global default instead
of the caller timeout. Add a per-request timeout to get() and forward the
configured timeout from the Soniox handler.

* fix(soniox): price stt-async-v4 per second so transcriptions are billed

The handler stores audio_transcription_duration in _hidden_params, but the
model carried only token cost fields and the response has no token usage, so
the transcription cost path fell through to cost_per_second and returned $0.
An authenticated caller could transcribe Soniox audio without decrementing
their budget. Switch the entry to output_cost_per_second at Soniox's published
$0.10/hour async rate so the stored duration produces a real charge.

* fix(langfuse): use a dedicated httpx client for the SDK injection

The httpx_client handed to the Langfuse SDK came from _get_httpx_client(),
which returns LiteLLM's globally cached HTTPHandler. If Langfuse closed that
client on teardown it would invalidate the shared client used by every other
LiteLLM HTTP call. Build a dedicated httpx.Client instead, still resolving SSL
verification and client certificate from LiteLLM's configuration.

* fix(soniox): prefer caller-supplied api_base over SONIOX_API_BASE env var

* fix(cohere): support max_completion_tokens on cohere v2 chat (default route) (BerriAI#29779)

* fix(cohere): support max_completion_tokens on cohere v2 chat

The default cohere_chat route resolves to CohereV2ChatConfig, which did not
list or map max_completion_tokens, so get_optional_params raised
UnsupportedParamsError for the standard OpenAI parameter (the modern
replacement for the deprecated max_tokens). The v1 config already maps it to
cohere's max_tokens; mirror that in v2 and add v2 regression tests.

* fix(cohere): make max_completion_tokens take precedence over max_tokens on v2

When both max_tokens and max_completion_tokens are supplied, prefer
max_completion_tokens explicitly rather than relying on dict iteration order,
and cover both orderings with a regression test.

---------

Co-authored-by: Daniel Yudelevich <4537920+yudelevi@users.noreply.github.com>
Co-authored-by: hectorc98 <hector.chamorroalvarez@adyen.com>
Co-authored-by: Filippo Menghi <113345637+Cyberfilo@users.noreply.github.com>
Co-authored-by: Terrajlz <info@jouleselectrictech.com>
Co-authored-by: Bruno Devaux <devaux.br@gmail.com>
Co-authored-by: Dan Lemon <dan@danlemon.com>
Co-authored-by: Saswat <saswatds@users.noreply.github.com>
Co-authored-by: Brian Sparker <brainsparker@users.noreply.github.com>
Co-authored-by: Zhao73 <156770117+Zhao73@users.noreply.github.com>
Co-authored-by: Urain Ahmad Shah <60431964+urainshah@users.noreply.github.com>
Co-authored-by: shin-berri <shin-laptop@berri.ai>
Co-authored-by: yuneng-jiang <yuneng@berri.ai>
Co-authored-by: kape <168134658+kapelame@users.noreply.github.com>
Co-authored-by: danisalvaa <159898202+danisalvaa@users.noreply.github.com>
Co-authored-by: Just R <remixingmagelang@gmail.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Co-authored-by: abhay23-AI <abhaytrivedi22@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants