On pre_call, during_call and post_call the hook data is the client request body, so a
client-supplied litellm_params.metadata.internal_call_origin must not skip enforcement.
Type the during_call helper's request payload as dict[str, object]
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
The request-side prompt told the judge to focus on the most recent user turn but the text under review was every extracted request message joined together, so a multi-turn request with an off-topic earlier turn and an on-topic latest turn scored 50 and was blocked. Request-side judging now evaluates the last extracted request text (after the configured message scoping) and passes the full role-labelled conversation only as context. Response-side judging still evaluates all extracted response text
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Judge sub-calls now carry the internal_call_origin metadata stamp and the
guardrail skips any logged call bearing it, so a logging_only judge no longer
recurses into an unbounded chain of judge requests
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Nova InvokeModel rejects the standalone cachePoint blocks the shared Converse transform emits, so each one is folded into the block it caches and tool_config injection points are dropped before the transform runs, since this route has no tool caching to credit. Usage reads Bedrock's Count-suffixed cache keys and adds cached tokens into prompt_tokens, streaming routes every wrapped InvokeModel event through the Converse chunk parser and tolerates the missing totalTokens, and the Nova 1 cost-map entries gain cache_read_input_token_cost at a quarter of the input rate
The /anthropic/{endpoint} route forwarded every incoming header upstream, so
the header carrying the caller's LiteLLM virtual key (Authorization, x-api-key,
x-litellm-api-key, or the operator-configured key header) reached Anthropic and
was rejected there as an invalid credential, with or without a proxy-side
Anthropic key layered on top.
Share the Vertex credential-less header filter: drop the proxy-only credential
headers by name, drop the value that authenticated the caller (virtual key,
master key, or JWT) from Authorization / x-api-key, keep a caller's own
Anthropic credential, layer the proxy's Anthropic credential on top, and fail
with a clean 401 when neither the proxy nor the caller supplied one.
Resolves LIT-3550
A team can now carry a per-model budget map that every key on the team
inherits. A key's own model_max_budget entry for the same model takes
precedence, so it is gated on and billed to the key alone.
Backend: NewTeamRequest/UpdateTeamRequest accept model_max_budget (validated
like the key-level field, enterprise gated); the value is hydrated onto
UserAPIKeyAuth via the token view, TeamGrants and the carried budget state;
_check_team_model_budget enforces it in the centralized common checks; the
limiter meters spend under team_model_spend:<team>:<model>:<duration> and
skips the team counter when the key overrides; /team/update lets only a
proxy admin raise, re-window or drop a cap; /team/info exposes usage.
The Anthropic context-management compaction summary subrequest runs the
same team gate. Both fallback token-view SQL definitions project the column.
UI: team create and edit forms reuse the key-level ModelMaxBudgetEditor,
premium gated, sending {} to clear and omitting unchanged fields.
A key entry overrides the team cap only when it spend-gates the model
(non-negative max_budget); a row that only carries tpm/rpm limits or a
negative cap leaves the team cap in force.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
