> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/relay/_mcp/server.

# Module llm

> LLM lifecycle helpers and managed execution entry points.

Generated from `cargo doc --no-deps -p nemo-relay -p nemo-relay-adaptive -p nemo-relay-pii-redaction -p nemo-relay-ffi -p nemo-relay-types -p nemo-relay-plugin -p nemo-relay-worker-proto -p nemo-relay-worker`.

LLM lifecycle helpers and managed execution entry points.

## Structs

* [CreateLlmHandleParams](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-createllmhandleparams): Builder parameters for [`NemoRelayContextState::create_llm_handle`](/reference/api/rust-library-reference/nemo-relay/api/runtime/state/struct-nemorelaycontextstate).
* [EndLlmHandleParams](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-endllmhandleparams): Builder parameters for [`NemoRelayContextState::build_llm_end_event`](/reference/api/rust-library-reference/nemo-relay/api/runtime/state/struct-nemorelaycontextstate).
* [LlmAttributes](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmattributes): Bitflags that modify LLM-call behavior and observability.
* [LlmCallEndParams](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmcallendparams): Builder parameters for [`llm_call_end`](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call-end).
* [LlmCallExecuteParams](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmcallexecuteparams): Builder parameters for [`llm_call_execute`](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call-execute).
* [LlmCallParams](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmcallparams): Builder parameters for [`llm_call`](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call).
* [LlmHandle](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmhandle): Runtime-owned handle identifying an active or completed LLM call.
* [LlmRequest](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmrequest): JSON-shaped LLM request payload passed through the runtime.
* [LlmRequestInterceptOutcome](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmrequestinterceptoutcome): Result of an LLM request intercept that can schedule lifecycle marks.
* [LlmStreamCallExecuteParams](/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmstreamcallexecuteparams): Builder parameters for [`llm_stream_call_execute`](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-stream-call-execute).

## Functions

* [llm\_call](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call): Start a manual LLM lifecycle span.
* [llm\_call\_end](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call-end): Finish a manual LLM lifecycle span.
* [llm\_call\_execute](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call-execute): Execute an LLM call through the managed middleware pipeline.
* [llm\_conditional\_execution](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-conditional-execution): Run only the LLM conditional-execution guardrail chain.
* [llm\_request\_intercepts](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-request-intercepts): Run only the LLM request-intercept chain.
* [llm\_stream\_call\_execute](/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-stream-call-execute): Execute a streaming LLM call through the managed middleware pipeline.

> LLM lifecycle helpers and managed execution entry points.

## Docs

- [Struct Create LlmHandle Params](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-createllmhandleparams.md): Builder parameters for `NemoRelayContextState::create_llm_handle`.
- [Struct EndLlm Handle Params](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-endllmhandleparams.md): Builder parameters for `NemoRelayContextState::build_llm_end_event`.
- [Struct LlmAttributes](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmattributes.md): Bitflags that modify LLM-call behavior and observability.
- [Struct LlmCall EndParams](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmcallendparams.md): Builder parameters for `llm_call_end`.
- [Struct LlmCall Execute Params](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmcallexecuteparams.md): Builder parameters for `llm_call_execute`.
- [Struct LlmCall Params](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmcallparams.md): Builder parameters for `llm_call`.
- [Struct LlmHandle](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmhandle.md): Runtime-owned handle identifying an active or completed LLM call.
- [Struct LlmRequest](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmrequest.md): JSON-shaped LLM request payload passed through the runtime.
- [Struct LlmRequest Intercept Outcome](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmrequestinterceptoutcome.md): Result of an LLM request intercept that can schedule lifecycle marks.
- [Struct LlmStream Call Execute Params](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/struct-llmstreamcallexecuteparams.md): Builder parameters for `llm_stream_call_execute`.
- [Function llm_call](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call.md): Start a manual LLM lifecycle span.
- [Function llm_call_end](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call-end.md): Finish a manual LLM lifecycle span.
- [Function llm_call_execute](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-call-execute.md): Execute an LLM call through the managed middleware pipeline.
- [Function llm_conditional_execution](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-conditional-execution.md): Run only the LLM conditional-execution guardrail chain.
- [Function llm_request_intercepts](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-request-intercepts.md): Run only the LLM request-intercept chain.
- [Function llm_stream_call_execute](https://docs.nvidia.com/nemo/relay/v0.5.0/reference/api/rust-library-reference/nemo-relay/api/llm/fn-llm-stream-call-execute.md): Execute a streaming LLM call through the managed middleware pipeline.