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

# Function nemo_relay_register_llm_stream_execution_intercept

> Register an LLM streaming execution intercept following the middleware chain pattern. The callback receives `(request, next_fn, next_ctx)` - call `next_fn(request, next_ctx)` to invoke the next intercept or the original streaming LLM call, or skip calling it to short-circuit.

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`.

<pre />

Register an LLM streaming execution intercept following the middleware chain pattern. The callback receives `(request, next_fn, next_ctx)` - call `next_fn(request, next_ctx)` to invoke the next intercept or the original streaming LLM call, or skip calling it to short-circuit.

## Parameters

* `name`: Unique intercept name.
* `priority`: Execution priority (lower runs first).
* `exec_cb`: Middleware callback receiving request and a next function.
* `exec_user_data`: Opaque pointer for the execution callback.
* `exec_free`: Optional destructor for `exec_user_data`.

## Safety

`name` must be a valid C string. Callback pointers must be valid.