> 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.

# Struct LlmOptimization Tokens

> Token quantities retained independently from pricing arithmetic.

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 />

Token quantities retained independently from pricing arithmetic.

## Fields

### `prompt_tokens: Option<u64>`

Input/prompt tokens.

### `completion_tokens: Option<u64>`

Output/completion tokens.

### `cache_read_tokens: Option<u64>`

Tokens read from a provider prompt cache.

### `cache_write_tokens: Option<u64>`

Tokens written to a provider prompt cache.

### `total_tokens: Option<u64>`

Total tokens when supplied by the producer.

## Implementations

### `impl LlmOptimizationTokens`

<pre />

#### `saved_prompt`

<pre />

Construct explicit prompt and total token savings.

## Trait Implementations

### `impl Clone for LlmOptimizationTokens`

<pre />

#### `clone`

<pre />

#### `clone_from`

<pre />

### `impl Debug for LlmOptimizationTokens`

<pre />

#### `fmt`

<pre />

### `impl Default for LlmOptimizationTokens`

<pre />

#### `default`

<pre />

### `impl<'de> Deserialize<'de> for LlmOptimizationTokens`

<pre />

#### `deserialize`

<pre />

### `impl PartialEq for LlmOptimizationTokens`

<pre />

#### `eq`

<pre />

#### `ne`

<pre />

### `impl Serialize for LlmOptimizationTokens`

<pre />

#### `serialize`

<pre />

### `impl Eq for LlmOptimizationTokens`

<pre />

### `impl StructuralPartialEq for LlmOptimizationTokens`

<pre />