# @arizeai/openinference-semantic-conventions

**JavaScript · Core**

This package provides OpenInference semantic conventions for tracing of LLM Applications.

- **Version**: 2.17.1

[npm](https://www.npmjs.com/package/@arizeai/openinference-semantic-conventions) · [README](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/README.md) · [Source](https://github.com/Arize-ai/openinference/tree/main/js/packages/openinference-semantic-conventions)

## Install

```sh
npm install @arizeai/openinference-semantic-conventions
```

## `@arizeai/openinference-semantic-conventions`

### `SemanticAttributePrefixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L5)

```typescript
const SemanticAttributePrefixes
```

Semantic conventions for OpenInference tracing

**Properties**

| Name | Type |
| - | - |
| [`annotations`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.annotations) | `"annotations"` |
| [`annotation`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.annotation) | `"annotation"` |
| [`evaluations`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.evaluations) | `"evaluations"` |
| [`evaluation`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.evaluation) | `"evaluation"` |
| [`input`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.input) | `"input"` |
| [`output`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.output) | `"output"` |
| [`llm`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.llm) | `"llm"` |
| [`retrieval`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.retrieval) | `"retrieval"` |
| [`reranker`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.reranker) | `"reranker"` |
| [`messages`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.messages) | `"messages"` |
| [`message`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.message) | `"message"` |
| [`document`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.document) | `"document"` |
| [`embedding`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.embedding) | `"embedding"` |
| [`tool`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.tool) | `"tool"` |
| [`tool_call`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.tool_call) | `"tool_call"` |
| [`metadata`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.metadata) | `"metadata"` |
| [`tag`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.tag) | `"tag"` |
| [`session`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.session) | `"session"` |
| [`trace`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.trace) | `"trace"` |
| [`user`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.user) | `"user"` |
| [`openinference`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.openinference) | `"openinference"` |
| [`message_content`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.message_content) | `"message_content"` |
| [`image`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.image) | `"image"` |
| [`audio`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.audio) | `"audio"` |
| [`video`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.video) | `"video"` |
| [`prompt`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.prompt) | `"prompt"` |
| [`agent`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.agent) | `"agent"` |
| [`graph`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.graph) | `"graph"` |
| [`decision`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticAttributePrefixes.decision) | `"decision"` |

### `LLMAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L37)

```typescript
const LLMAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`provider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.provider) | `"provider"` |
| [`system`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.system) | `"system"` |
| [`model_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.model_name) | `"model_name"` |
| [`request`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.request) | `"request"` |
| [`response`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.response) | `"response"` |
| [`token_count`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.token_count) | `"token_count"` |
| [`input_messages`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.input_messages) | `"input_messages"` |
| [`output_messages`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.output_messages) | `"output_messages"` |
| [`invocation_parameters`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.invocation_parameters) | `"invocation_parameters"` |
| [`prompts`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.prompts) | `"prompts"` |
| [`prompt_template`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.prompt_template) | `"prompt_template"` |
| [`function_call`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.function_call) | `"function_call"` |
| [`tools`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.tools) | `"tools"` |
| [`cost`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.cost) | `"cost"` |
| [`finish_reason`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMAttributePostfixes.finish_reason) | `"finish_reason"` |

### `DecisionAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L55)

```typescript
const DecisionAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`provider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionAttributePostfixes.provider) | `"provider"` |
| [`system`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionAttributePostfixes.system) | `"system"` |
| [`model_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionAttributePostfixes.model_name) | `"model_name"` |
| [`request`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionAttributePostfixes.request) | `"request"` |
| [`response`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionAttributePostfixes.response) | `"response"` |
| [`token_count`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionAttributePostfixes.token_count) | `"token_count"` |

### `LLMPromptTemplateAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L64)

```typescript
const LLMPromptTemplateAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`variables`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMPromptTemplateAttributePostfixes.variables) | `"variables"` |
| [`template`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMPromptTemplateAttributePostfixes.template) | `"template"` |

### `RetrievalAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L69)

```typescript
const RetrievalAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`documents`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RetrievalAttributePostfixes.documents) | `"documents"` |

### `RerankerAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L73)

```typescript
const RerankerAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`input_documents`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RerankerAttributePostfixes.input_documents) | `"input_documents"` |
| [`output_documents`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RerankerAttributePostfixes.output_documents) | `"output_documents"` |
| [`query`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RerankerAttributePostfixes.query) | `"query"` |
| [`model_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RerankerAttributePostfixes.model_name) | `"model_name"` |
| [`top_k`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RerankerAttributePostfixes.top_k) | `"top_k"` |

### `EmbeddingAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L81)

```typescript
const EmbeddingAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`embeddings`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EmbeddingAttributePostfixes.embeddings) | `"embeddings"` |
| [`text`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EmbeddingAttributePostfixes.text) | `"text"` |
| [`model_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EmbeddingAttributePostfixes.model_name) | `"model_name"` |
| [`vector`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EmbeddingAttributePostfixes.vector) | `"vector"` |

### `ToolAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L88)

```typescript
const ToolAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolAttributePostfixes.name) | `"name"` |
| [`description`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolAttributePostfixes.description) | `"description"` |
| [`parameters`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolAttributePostfixes.parameters) | `"parameters"` |
| [`json_schema`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolAttributePostfixes.json_schema) | `"json_schema"` |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolAttributePostfixes.id) | `"id"` |

### `MessageAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L96)

```typescript
const MessageAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`role`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.role) | `"role"` |
| [`content`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.content) | `"content"` |
| [`contents`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.contents) | `"contents"` |
| [`name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.name) | `"name"` |
| [`function_call_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.function_call_name) | `"function_call_name"` |
| [`function_call_arguments_json`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.function_call_arguments_json) | `"function_call_arguments_json"` |
| [`tool_calls`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.tool_calls) | `"tool_calls"` |
| [`tool_call_id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageAttributePostfixes.tool_call_id) | `"tool_call_id"` |

### `MessageContentsAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L107)

```typescript
const MessageContentsAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`type`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.type) | `"type"` |
| [`text`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.text) | `"text"` |
| [`image`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.image) | `"image"` |
| [`audio`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.audio) | `"audio"` |
| [`video`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.video) | `"video"` |
| [`document`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.document) | `"document"` |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.id) | `"id"` |
| [`signature`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.signature) | `"signature"` |
| [`data`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.data) | `"data"` |
| [`encrypted_content`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MessageContentsAttributePostfixes.encrypted_content) | `"encrypted_content"` |

### `ImageAttributesPostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L120)

```typescript
const ImageAttributesPostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`url`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ImageAttributesPostfixes.url) | `"url"` |

### `VideoAttributesPostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L124)

```typescript
const VideoAttributesPostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`url`](https://openinf.dev/js/openinference-semantic-conventions.html.md#VideoAttributesPostfixes.url) | `"url"` |

### `ToolCallAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L128)

```typescript
const ToolCallAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`function_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolCallAttributePostfixes.function_name) | `"function.name"` |
| [`function_arguments_json`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolCallAttributePostfixes.function_arguments_json) | `"function.arguments"` |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolCallAttributePostfixes.id) | `"id"` |
| [`reasoning_signature`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ToolCallAttributePostfixes.reasoning_signature) | `"reasoning_signature"` |

### `DocumentAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L135)

```typescript
const DocumentAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.id) | `"id"` |
| [`content`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.content) | `"content"` |
| [`score`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.score) | `"score"` |
| [`metadata`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.metadata) | `"metadata"` |
| [`url`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.url) | `"url"` |
| [`mime_type`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.mime_type) | `"mime_type"` |
| [`name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DocumentAttributePostfixes.name) | `"name"` |

### `TagAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L145)

```typescript
const TagAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`tags`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TagAttributePostfixes.tags) | `"tags"` |

### `SessionAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L149)

```typescript
const SessionAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SessionAttributePostfixes.id) | `"id"` |

### `UserAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L153)

```typescript
const UserAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#UserAttributePostfixes.id) | `"id"` |

### `AudioAttributesPostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L157)

```typescript
const AudioAttributesPostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`url`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AudioAttributesPostfixes.url) | `"url"` |
| [`mime_type`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AudioAttributesPostfixes.mime_type) | `"mime_type"` |
| [`transcript`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AudioAttributesPostfixes.transcript) | `"transcript"` |

### `PromptAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L163)

```typescript
const PromptAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`vendor`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PromptAttributePostfixes.vendor) | `"vendor"` |
| [`id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PromptAttributePostfixes.id) | `"id"` |
| [`url`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PromptAttributePostfixes.url) | `"url"` |

### `AgentPostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L169)

```typescript
const AgentPostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AgentPostfixes.name) | `"name"` |

### `GraphPostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L173)

```typescript
const GraphPostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`node_id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#GraphPostfixes.node_id) | `"node.id"` |
| [`node_name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#GraphPostfixes.node_name) | `"node.name"` |
| [`node_parent_id`](https://openinf.dev/js/openinference-semantic-conventions.html.md#GraphPostfixes.node_parent_id) | `"node.parent_id"` |

### `FeedbackAttributePostfixes`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L179)

```typescript
const FeedbackAttributePostfixes
```

**Properties**

| Name | Type |
| - | - |
| [`name`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.name) | `"name"` |
| [`score`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.score) | `"score"` |
| [`label`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.label) | `"label"` |
| [`explanation`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.explanation) | `"explanation"` |
| [`annotator_kind`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.annotator_kind) | `"annotator_kind"` |
| [`identifier`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.identifier) | `"identifier"` |
| [`metadata`](https://openinf.dev/js/openinference-semantic-conventions.html.md#FeedbackAttributePostfixes.metadata) | `"metadata"` |

### `SemanticConventions`

_constant_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L889)

```typescript
const SemanticConventions
```

**Properties**

| Name | Type |
| - | - |
| [`ANNOTATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATIONS) | `"annotations"` |
| [`TRACE_ANNOTATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TRACE_ANNOTATIONS) | `"trace.annotations"` |
| [`SESSION_ANNOTATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.SESSION_ANNOTATIONS) | `"session.annotations"` |
| [`ANNOTATION_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_NAME) | `"annotation.name"` |
| [`ANNOTATION_SCORE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_SCORE) | `"annotation.score"` |
| [`ANNOTATION_LABEL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_LABEL) | `"annotation.label"` |
| [`ANNOTATION_EXPLANATION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_EXPLANATION) | `"annotation.explanation"` |
| [`ANNOTATION_ANNOTATOR_KIND`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_ANNOTATOR_KIND) | `"annotation.annotator_kind"` |
| [`ANNOTATION_IDENTIFIER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_IDENTIFIER) | `"annotation.identifier"` |
| [`ANNOTATION_METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.ANNOTATION_METADATA) | `"annotation.metadata"` |
| [`EVALUATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATIONS) | `"evaluations"` |
| [`TRACE_EVALUATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TRACE_EVALUATIONS) | `"trace.evaluations"` |
| [`SESSION_EVALUATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.SESSION_EVALUATIONS) | `"session.evaluations"` |
| [`EVALUATION_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_NAME) | `"evaluation.name"` |
| [`EVALUATION_SCORE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_SCORE) | `"evaluation.score"` |
| [`EVALUATION_LABEL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_LABEL) | `"evaluation.label"` |
| [`EVALUATION_EXPLANATION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_EXPLANATION) | `"evaluation.explanation"` |
| [`EVALUATION_ANNOTATOR_KIND`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_ANNOTATOR_KIND) | `"evaluation.annotator_kind"` |
| [`EVALUATION_IDENTIFIER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_IDENTIFIER) | `"evaluation.identifier"` |
| [`EVALUATION_METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EVALUATION_METADATA) | `"evaluation.metadata"` |
| [`IMAGE_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.IMAGE_URL) | `"image.url"`The http or base64 link to the image |
| [`VIDEO_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.VIDEO_URL) | `"video.url"`The URL, object-store URI, or base64 data URI of a video. |
| [`INPUT_VALUE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.INPUT_VALUE) | `"input.value"`The input to any span |
| [`INPUT_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.INPUT_MIME_TYPE) | `"input.mime_type"` |
| [`INPUT_IMAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.INPUT_IMAGES) | `"input.images"`Images passed as input to a span of any kind. Flattened with an index, for example `input.images.0.image.url`. |
| [`INPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.INPUT_DOCUMENTS) | `"input.documents"`Documents (e.g. PDFs) passed as input to a span of any kind. Flattened with an index, for example `input.documents.0.document.url`. |
| [`OUTPUT_VALUE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.OUTPUT_VALUE) | `"output.value"`The output of any span |
| [`OUTPUT_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.OUTPUT_MIME_TYPE) | `"output.mime_type"` |
| [`OUTPUT_IMAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.OUTPUT_IMAGES) | `"output.images"`Images produced as output by a span of any kind. Flattened with an index, for example `output.images.0.image.url`. |
| [`OUTPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.OUTPUT_DOCUMENTS) | `"output.documents"`Documents (e.g. PDFs) produced as output by a span of any kind. Flattened with an index, for example `output.documents.0.document.url`. |
| [`LLM_INPUT_MESSAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_INPUT_MESSAGES) | `"llm.input_messages"`The messages sent to the LLM for completions Typically seen in OpenAI chat completions> **See also**
>
> <https://beta.openai.com/docs/api-reference/completions/create> |
| [`LLM_OUTPUT_MESSAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_OUTPUT_MESSAGES) | `"llm.output_messages"`The messages received from the LLM for completions Typically seen in OpenAI chat completions> **See also**
>
> <https://platform.openai.com/docs/api-reference/chat/object#choices-message> |
| [`LLM_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_MODEL_NAME) | `"llm.model_name"`The name of the LLM model |
| [`LLM_REQUEST_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_REQUEST_MODEL_NAME) | `"llm.request.model_name"`The model requested by the caller, as sent in the request. May differ from llm.response.model_name when the provider routes the request to a different model (e.g. classifier-triggered fallback). |
| [`LLM_RESPONSE_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_RESPONSE_MODEL_NAME) | `"llm.response.model_name"`The model that actually generated the response, as reported by the provider. May differ from llm.request.model_name when the provider routes the request to a different model (e.g. classifier-triggered fallback). |
| [`LLM_PROMPTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_PROMPTS) | `"llm.prompts"`The prompts sent to the LLM for completions Typically seen in OpenAI legacy completions> **See also**
>
> <https://beta.openai.com/docs/api-reference/completions/create> |
| [`LLM_INVOCATION_PARAMETERS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_INVOCATION_PARAMETERS) | `"llm.invocation_parameters"`The JSON representation of the parameters passed to the LLM |
| [`LLM_TOKEN_COUNT_COMPLETION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_COMPLETION) | `"llm.token_count.completion"`Token count for the completion by the llm (in tokens) |
| [`LLM_TOKEN_COUNT_COMPLETION_DETAILS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_COMPLETION_DETAILS) | `"llm.token_count.completion_details"`Key prefix for additional completion token count details. Each detail should be a separate attribute with this prefix, e.g. llm.token_count.completion_details.reasoning, llm.token_count.completion_details.audio. All values should be in tokens (integer count of tokens). |
| [`LLM_TOKEN_COUNT_COMPLETION_DETAILS_REASONING`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_COMPLETION_DETAILS_REASONING) | `"llm.token_count.completion_details.reasoning"`Token count for the reasoning steps in the completion (in tokens) |
| [`LLM_TOKEN_COUNT_COMPLETION_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_COMPLETION_DETAILS_AUDIO) | `"llm.token_count.completion_details.audio"`Token count for audio input generated by the model (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_PROMPT) | `"llm.token_count.prompt"`Token count for the prompt to the llm (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_PROMPT_DETAILS) | `"llm.token_count.prompt_details"`Key prefix for additional prompt token count details. Each detail should be a separate attribute with this prefix, e.g. llm.token_count.prompt_details.reasoning, llm.token_count.prompt_details.audio. All values should be in tokens (integer count of tokens). |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_WRITE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_WRITE) | `"llm.token_count.prompt_details.cache_write"`Token count for the tokens written to cache (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_READ`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_READ) | `"llm.token_count.prompt_details.cache_read"`Token count for the tokens retrieved from cache (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_INPUT) | `"llm.token_count.prompt_details.cache_input"`Token count for the input tokens in the prompt that were cached (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_PROMPT_DETAILS_AUDIO) | `"llm.token_count.prompt_details.audio"`Token count for audio input presented in the prompt (in tokens) |
| [`LLM_TOKEN_COUNT_TOTAL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOKEN_COUNT_TOTAL) | `"llm.token_count.total"`Token count for the entire transaction with the llm (in tokens) |
| [`LLM_SYSTEM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_SYSTEM) | `"llm.system"`The AI product as identified by the client or server |
| [`LLM_PROVIDER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_PROVIDER) | `"llm.provider"`The provider of the inferences. E.g. the cloud provider |
| [`DECISION_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_MODEL_NAME) | `"decision.model_name"`The name of the decision model. Mirrors LLM_MODEL_NAME for DECISION spans. A decision model takes state plus typed questions and returns a typed, probabilistic answer per question instead of generated text.> **See also**
>
> <https://github.com/Arize-ai/openinference/blob/main/spec/spans/decision.md> |
| [`DECISION_REQUEST_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_REQUEST_MODEL_NAME) | `"decision.request.model_name"`The decision model requested by the caller, as sent in the request. May differ from decision.response.model_name when the provider resolves an alias (e.g. jev-latest) or routes the request to a different model. |
| [`DECISION_RESPONSE_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_RESPONSE_MODEL_NAME) | `"decision.response.model_name"`The decision model that actually produced the response, as reported by the provider. May differ from decision.request.model_name. |
| [`DECISION_SYSTEM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_SYSTEM) | `"decision.system"`The decision API ecosystem the call conforms to, i.e. which request and response shape the client speaks, as identified by the client or server: “typesafe” for the TypeSafe System One / Jev API, “openai” for the OpenAI Decisions API. A self-hosted vLLM server answering the Jev-compatible /v1/systemone shape is still “typesafe”. Distinct from DECISION_PROVIDER, which says who hosts the model. Well-known values are [`DecisionSystem`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionSystem), which alias the matching [`LLMSystem`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem) values. |
| [`DECISION_PROVIDER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_PROVIDER) | `"decision.provider"`The hosting provider of the decision model: who runs the hardware that answered (e.g. “typesafe” when calling TypeSafe directly, or the cloud or self-hosting provider for a model served elsewhere). Distinct from DECISION_SYSTEM, which names the API shape. Well-known values are [`DecisionProvider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionProvider), which alias the matching [`LLMProvider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider) values. |
| [`DECISION_TOKEN_COUNT_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_TOKEN_COUNT_INPUT) | `"decision.token_count.input"`The number of input tokens consumed by a decision model call: the state, questions, and candidate options. Maps to usage.input_tokens in TypeSafe System One responses. |
| [`DECISION_TOKEN_COUNT_OUTPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DECISION_TOKEN_COUNT_OUTPUT) | `"decision.token_count.output"`The number of output tokens produced by a decision model call. Decision models emit typed answers rather than text, so this is typically small. Maps to usage.output_tokens in TypeSafe System One responses. |
| [`LLM_TOOLS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_TOOLS) | `"llm.tools"`List of tools that are advertised to the LLM to be able to call |
| [`LLM_FINISH_REASON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_FINISH_REASON) | `"llm.finish_reason"`The reason the model stopped generating tokens, e.g. `"stop"` or `"length"`. |
| [`LLM_COST`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST) | `"llm.cost"`Key prefix for cost information. When these keys are transformed into a JSON-like structure, it would look like: { “prompt”: 0.0021, # Cost in USD “completion”: 0.0045, # Cost in USD “total”: 0.0066, # Cost in USD “completion_details”: { “output”: 0.0009, # Cost in USD “reasoning”: 0.0024, # Cost in USD (e.g., 80 tokens \* $0.03/1K tokens) “audio”: 0.0012 # Cost in USD (e.g., 40 tokens \* $0.03/1K tokens) }, “prompt_details”: { “input”: 0.0003, # Cost in USD “cache_write”: 0.0006, # Cost in USD (e.g., 20 tokens \* $0.03/1K tokens) “cache_read”: 0.0003, # Cost in USD (e.g., 10 tokens \* $0.03/1K tokens) “cache_input”: 0.0006, # Cost in USD (e.g., 20 tokens \* $0.03/1K tokens) “audio”: 0.0003 # Cost in USD (e.g., 10 tokens \* $0.03/1K tokens) } } Note: This is a key prefix - individual attributes are stored as separate span attributes with this prefix, e.g. llm.cost.prompt, llm.cost.completion_details.reasoning, etc. The JSON structure shown above represents how these separate attributes can be conceptually organized. All monetary values are in USD with floating point precision. |
| [`LLM_COST_PROMPT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_PROMPT) | `"llm.cost.prompt"`Cost of the prompt tokens in USD |
| [`LLM_COST_COMPLETION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_COMPLETION) | `"llm.cost.completion"`Cost of the completion tokens in USD |
| [`LLM_COST_TOTAL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_TOTAL) | `"llm.cost.total"`Total cost of the LLM call in USD (prompt + completion) |
| [`LLM_COST_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_INPUT) | `"llm.cost.prompt_details.input"`Total cost of input tokens in USD. This represents the cost of tokens that were used as input to the model, which may be different from the prompt cost if there are additional processing steps. |
| [`LLM_COST_OUTPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_OUTPUT) | `"llm.cost.completion_details.output"`Total cost of output tokens in USD. This represents the cost of tokens that were generated as output by the model, which may be different from the completion cost if there are additional processing steps. |
| [`LLM_COST_COMPLETION_DETAILS_REASONING`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_COMPLETION_DETAILS_REASONING) | `"llm.cost.completion_details.reasoning"`Cost of reasoning steps in the completion in USD |
| [`LLM_COST_COMPLETION_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_COMPLETION_DETAILS_AUDIO) | `"llm.cost.completion_details.audio"`Cost of audio tokens in the completion in USD |
| [`LLM_COST_PROMPT_DETAILS_CACHE_WRITE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_PROMPT_DETAILS_CACHE_WRITE) | `"llm.cost.prompt_details.cache_write"`Cost of prompt tokens written to cache in USD |
| [`LLM_COST_PROMPT_DETAILS_CACHE_READ`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_PROMPT_DETAILS_CACHE_READ) | `"llm.cost.prompt_details.cache_read"`Cost of prompt tokens read from cache in USD |
| [`LLM_COST_PROMPT_DETAILS_CACHE_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_PROMPT_DETAILS_CACHE_INPUT) | `"llm.cost.prompt_details.cache_input"`Cost of input tokens in the prompt that were cached in USD |
| [`LLM_COST_PROMPT_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_COST_PROMPT_DETAILS_AUDIO) | `"llm.cost.prompt_details.audio"`Cost of audio tokens in the prompt in USD |
| [`MESSAGE_ROLE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_ROLE) | `"message.role"`The role that the LLM assumes the message is from during the LLM invocation |
| [`MESSAGE_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_NAME) | `"message.name"`The name of the message. This is only used for role ‘function’ where the name of the function is captured in the name field and the parameters are captured in the content. |
| [`MESSAGE_TOOL_CALLS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_TOOL_CALLS) | `"message.tool_calls"`The tool calls generated by the model, such as function calls. |
| [`MESSAGE_TOOL_CALL_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_TOOL_CALL_ID) | `"message.tool_call_id"`The id of the tool call on a “tool” role message |
| [`TOOL_CALL_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_CALL_ID) | `"tool_call.id"`The id of the tool call |
| [`TOOL_CALL_REASONING_SIGNATURE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_CALL_REASONING_SIGNATURE) | `"tool_call.reasoning_signature"`Opaque vendor-issued reasoning echo token attached to a tool call. Maps to Gemini thoughtSignature when it is attached to a functionCall part |
| [`TOOL_CALL_FUNCTION_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_CALL_FUNCTION_NAME) | `"tool_call.function.name"`tool_call.function.name |
| [`TOOL_CALL_FUNCTION_ARGUMENTS_JSON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_CALL_FUNCTION_ARGUMENTS_JSON) | `"tool_call.function.arguments"`tool_call.function.argument (JSON string) |
| [`MESSAGE_FUNCTION_CALL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_FUNCTION_CALL_NAME) | `"message.function_call_name"`The LLM function call function name |
| [`MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON) | `"message.function_call_arguments_json"`The LLM function call function arguments in a json string |
| [`MESSAGE_CONTENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT) | `"message.content"`The content of the message sent to the LLM |
| [`MESSAGE_CONTENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENTS) | `"message.contents"`The array of contents for the message sent to the LLM. Each element of the array is an `message_content` object. |
| [`MESSAGE_CONTENT_IMAGE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_IMAGE) | `"message_content.image"`The image content of the message sent to the LLM |
| [`MESSAGE_CONTENT_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_AUDIO) | `"message_content.audio"`The audio content of the message sent to the LLM |
| [`MESSAGE_CONTENT_VIDEO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_VIDEO) | `"message_content.video"`The video content of the message sent to the LLM |
| [`MESSAGE_CONTENT_DOCUMENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_DOCUMENT) | `"message_content.document"`The document content (e.g. a PDF, an uploaded file, or inline text) of the message sent to the LLM |
| [`MESSAGE_CONTENT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_ID) | `"message_content.id"`Provider-assigned identifier for this message content item. For OpenAI Responses reasoning items, this maps to ResponseReasoningItem.id and should be preserved for stateless replay |
| [`MESSAGE_CONTENT_SIGNATURE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_SIGNATURE) | `"message_content.signature"`Opaque vendor-issued signature captured verbatim. Maps to provider signature fields and to Gemini thoughtSignature fields when the signature is attached to a non-tool content part |
| [`MESSAGE_CONTENT_DATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_DATA) | `"message_content.data"`Opaque vendor-issued data captured verbatim. Maps to Anthropic redacted_thinking.data |
| [`MESSAGE_CONTENT_ENCRYPTED_CONTENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_ENCRYPTED_CONTENT) | `"message_content.encrypted_content"`OpenAI encrypted_content captured verbatim |
| [`MESSAGE_CONTENT_TEXT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_TEXT) | `"message_content.text"`The text content of the message sent to the LLM |
| [`MESSAGE_CONTENT_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.MESSAGE_CONTENT_TYPE) | `"message_content.type"`The type of content sent to the LLM, such as “text”, “image”, “audio”, “video”, “document”, “reasoning”, or “tool_use” |
| [`DOCUMENT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_ID) | `"document.id"` |
| [`DOCUMENT_CONTENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_CONTENT) | `"document.content"` |
| [`DOCUMENT_SCORE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_SCORE) | `"document.score"` |
| [`DOCUMENT_METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_METADATA) | `"document.metadata"` |
| [`DOCUMENT_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_URL) | `"document.url"`The URL, object-store URI, or base64 data URI of a file document such as a PDF. |
| [`DOCUMENT_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_MIME_TYPE) | `"document.mime_type"`The MIME type of a document, for example `application/pdf`. |
| [`DOCUMENT_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.DOCUMENT_NAME) | `"document.name"`The name supplied with a document, such as a file name or a title, for example `invoice.pdf`. Not a unique identifier. |
| [`EMBEDDING_EMBEDDINGS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EMBEDDING_EMBEDDINGS) | `"embedding.embeddings"`The embedding list root |
| [`EMBEDDING_TEXT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EMBEDDING_TEXT) | `"embedding.text"`The text that was embedded to create the vector |
| [`EMBEDDING_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EMBEDDING_MODEL_NAME) | `"embedding.model_name"`The name of the model that was used to create the vector |
| [`EMBEDDING_VECTOR`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.EMBEDDING_VECTOR) | `"embedding.vector"`The embedding vector. Typically a high dimensional vector of floats or ints |
| [`TOOL_DESCRIPTION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_DESCRIPTION) | `"tool.description"`The description of a tool |
| [`TOOL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_NAME) | `"tool.name"`The name of a tool |
| [`TOOL_PARAMETERS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_PARAMETERS) | `"tool.parameters"`The parameters of the tool represented as a JSON string |
| [`TOOL_JSON_SCHEMA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_JSON_SCHEMA) | `"tool.json_schema"`The json schema of a tool input, It is RECOMMENDED that this be in the OpenAI tool calling format: <https://platform.openai.com/docs/assistants/tools> |
| [`TOOL_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TOOL_ID) | `"tool.id"`The identifier for the result of the tool call (corresponding to tool_call.id). |
| [`PROMPT_TEMPLATE_VARIABLES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.PROMPT_TEMPLATE_VARIABLES) | `"llm.prompt_template.variables"`The JSON representation of the variables used in the prompt template |
| [`PROMPT_TEMPLATE_TEMPLATE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.PROMPT_TEMPLATE_TEMPLATE) | `"llm.prompt_template.template"`A prompt template |
| [`PROMPT_TEMPLATE_VERSION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.PROMPT_TEMPLATE_VERSION) | `"llm.prompt_template.version"`A prompt template version |
| [`RERANKER_INPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.RERANKER_INPUT_DOCUMENTS) | `"reranker.input_documents"`The documents used as input to the reranker |
| [`RERANKER_OUTPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.RERANKER_OUTPUT_DOCUMENTS) | `"reranker.output_documents"`The documents output by the reranker |
| [`RERANKER_QUERY`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.RERANKER_QUERY) | `"reranker.query"`The query string for the reranker |
| [`RERANKER_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.RERANKER_MODEL_NAME) | `"reranker.model_name"`The model name for the reranker |
| [`RERANKER_TOP_K`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.RERANKER_TOP_K) | `"reranker.top_k"`The top k parameter for the reranker |
| [`LLM_FUNCTION_CALL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.LLM_FUNCTION_CALL) | `"llm.function_call"`The JSON representation of a function call of an LLM |
| [`RETRIEVAL_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.RETRIEVAL_DOCUMENTS) | `"retrieval.documents"`The retrieval documents list root |
| [`SESSION_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.SESSION_ID) | `"session.id"`The session id of a trace. Used to correlate spans in a single session. |
| [`USER_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.USER_ID) | `"user.id"`The user id of a trace. Used to correlate spans for a single user. |
| [`METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.METADATA) | `"metadata"`Metadata for a span, used to store user-defined key-value pairs |
| [`TAG_TAGS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.TAG_TAGS) | `"tag.tags"`The tags associated with a span |
| [`readonly OPENINFERENCE_SPAN_KIND`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.OPENINFERENCE_SPAN_KIND) | `"openinference.span.kind" = ...` |
| [`PROMPT_VENDOR`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.PROMPT_VENDOR) | `"prompt.vendor"`The vendor or origin of the prompt, e.g. a prompt library, a specialized service, etc. |
| [`PROMPT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.PROMPT_ID) | `"prompt.id"`A vendor-specific id used to locate the prompt |
| [`PROMPT_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.PROMPT_URL) | `"prompt.url"`A vendor-specific URL used to locate the prompt |
| [`AGENT_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.AGENT_NAME) | `"agent.name"`The name of the agent. Agents that perform the same functions should have the same name. |
| [`GRAPH_NODE_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.GRAPH_NODE_ID) | `"graph.node.id"`The id of the node in the execution graph. This along with graph.node.parent_id are used to visualize the execution graph. |
| [`GRAPH_NODE_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.GRAPH_NODE_NAME) | `"graph.node.name"`The name of the node in the execution graph. Use this to present a human readable name for the node. Optional |
| [`GRAPH_NODE_PARENT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SemanticConventions.GRAPH_NODE_PARENT_ID) | `"graph.node.parent_id"`This references the id of the parent node. Leaving this unset or set as empty string implies that the current span is the root node. |

### `AnnotatorKind`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1023)

```typescript
enum AnnotatorKind
```

**Members**

| Name | Value |
| - | - |
| [`HUMAN`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AnnotatorKind.HUMAN) | `"HUMAN"` |
| [`LLM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AnnotatorKind.LLM) | `"LLM"` |
| [`CODE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AnnotatorKind.CODE) | `"CODE"` |

### `OpenInferenceSpanKind`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1029)

```typescript
enum OpenInferenceSpanKind
```

**Members**

| Name | Value |
| - | - |
| [`LLM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.LLM) | `"LLM"` |
| [`CHAIN`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.CHAIN) | `"CHAIN"` |
| [`TOOL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.TOOL) | `"TOOL"` |
| [`RETRIEVER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.RETRIEVER) | `"RETRIEVER"` |
| [`RERANKER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.RERANKER) | `"RERANKER"` |
| [`EMBEDDING`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.EMBEDDING) | `"EMBEDDING"` |
| [`AGENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.AGENT) | `"AGENT"` |
| [`GUARDRAIL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.GUARDRAIL) | `"GUARDRAIL"` |
| [`EVALUATOR`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.EVALUATOR) | `"EVALUATOR"` |
| [`PROMPT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.PROMPT) | `"PROMPT"` |
| [`DECISION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OpenInferenceSpanKind.DECISION) | `"DECISION"` |

### `MimeType`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1046)

```typescript
enum MimeType
```

An enum of common mime types. Not exhaustive.

**Members**

| Name | Value |
| - | - |
| [`TEXT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MimeType.TEXT) | `"text/plain"` |
| [`JSON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MimeType.JSON) | `"application/json"` |
| [`AUDIO_WAV`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MimeType.AUDIO_WAV) | `"audio/wav"` |

### `LLMSystem`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1052)

```typescript
enum LLMSystem
```

**Members**

| Name | Value |
| - | - |
| [`OPENAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.OPENAI) | `"openai"` |
| [`ANTHROPIC`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.ANTHROPIC) | `"anthropic"` |
| [`MISTRALAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.MISTRALAI) | `"mistralai"` |
| [`COHERE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.COHERE) | `"cohere"` |
| [`VERTEXAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.VERTEXAI) | `"vertexai"` |
| [`AI21`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.AI21) | `"ai21"` |
| [`META`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.META) | `"meta"` |
| [`AMAZON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.AMAZON) | `"amazon"` |
| [`TYPESAFE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem.TYPESAFE) | `"typesafe"` |

### `LLMProvider`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1064)

```typescript
enum LLMProvider
```

**Members**

| Name | Value |
| - | - |
| [`OPENAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.OPENAI) | `"openai"` |
| [`ANTHROPIC`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.ANTHROPIC) | `"anthropic"` |
| [`MISTRALAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.MISTRALAI) | `"mistralai"` |
| [`COHERE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.COHERE) | `"cohere"` |
| [`GOOGLE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.GOOGLE) | `"google"` |
| [`AWS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.AWS) | `"aws"` |
| [`AZURE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.AZURE) | `"azure"` |
| [`XAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.XAI) | `"xai"` |
| [`DEEPSEEK`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.DEEPSEEK) | `"deepseek"` |
| [`GROQ`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.GROQ) | `"groq"` |
| [`FIREWORKS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.FIREWORKS) | `"fireworks"` |
| [`MOONSHOT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.MOONSHOT) | `"moonshot"` |
| [`CEREBRAS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.CEREBRAS) | `"cerebras"` |
| [`PERPLEXITY`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.PERPLEXITY) | `"perplexity"` |
| [`TOGETHER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.TOGETHER) | `"together"` |
| [`OLLAMA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.OLLAMA) | `"ollama"` |
| [`META`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.META) | `"meta"` |
| [`ZAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.ZAI) | `"zai"` |
| [`MINIMAX`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.MINIMAX) | `"minimax"` |
| [`TYPESAFE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.TYPESAFE) | `"typesafe"` |
| [`ORACLE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.ORACLE) | `"oracle"` |
| [`IBM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.IBM) | `"ibm"` |
| [`DOUBLEWORD`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider.DOUBLEWORD) | `"doubleword"` |

### `DecisionSystem`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1098)

```typescript
enum DecisionSystem
```

Well-known values for `decision.system`: the decision API ecosystem a DECISION span conforms to. Each member aliases the [`LLMSystem`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem) member for the same vendor, so the same string names the same vendor on LLM and DECISION spans. The list is the subset of vendors currently known to offer a decision API.

**Members**

| Name | Value |
| - | - |
| [`TYPESAFE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionSystem.TYPESAFE) | `"typesafe"`TypeSafe AI System One / Jev API, including Jev-compatible servers. |
| [`OPENAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionSystem.OPENAI) | `"openai"`OpenAI Decisions API. |

### `DecisionProvider`

_enum_ · [Source](https://github.com/Arize-ai/openinference/blob/main/js/packages/openinference-semantic-conventions/src/trace/SemanticConventions.ts#L1110)

```typescript
enum DecisionProvider
```

Well-known values for `decision.provider`: who hosts the decision model that answered. Each member aliases the [`LLMProvider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider) member for the same vendor.

**Members**

| Name | Value |
| - | - |
| [`TYPESAFE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionProvider.TYPESAFE) | `"typesafe"` |
| [`OPENAI`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionProvider.OPENAI) | `"openai"` |

**Constants**

| Name | Value |
| - | - |
| [`SEMRESATTRS_PROJECT_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SEMRESATTRS_PROJECT_NAME) | `"openinference.project.name" = PROJECT_NAME`The project name to group traces under for openinference compatible services |
| [`ANNOTATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATIONS) | `"annotations" = SemanticAttributePrefixes.annotations` |
| [`EVALUATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATIONS) | `"evaluations" = SemanticAttributePrefixes.evaluations` |
| [`TRACE_ANNOTATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TRACE_ANNOTATIONS) | `"trace.annotations"` |
| [`TRACE_EVALUATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TRACE_EVALUATIONS) | `"trace.evaluations"` |
| [`SESSION_ANNOTATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SESSION_ANNOTATIONS) | `"session.annotations"` |
| [`SESSION_EVALUATIONS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SESSION_EVALUATIONS) | `"session.evaluations"` |
| [`ANNOTATION_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_NAME) | `"annotation.name"` |
| [`ANNOTATION_SCORE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_SCORE) | `"annotation.score"` |
| [`ANNOTATION_LABEL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_LABEL) | `"annotation.label"` |
| [`ANNOTATION_EXPLANATION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_EXPLANATION) | `"annotation.explanation"` |
| [`ANNOTATION_ANNOTATOR_KIND`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_ANNOTATOR_KIND) | `"annotation.annotator_kind"` |
| [`ANNOTATION_IDENTIFIER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_IDENTIFIER) | `"annotation.identifier"` |
| [`ANNOTATION_METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#ANNOTATION_METADATA) | `"annotation.metadata"` |
| [`EVALUATION_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_NAME) | `"evaluation.name"` |
| [`EVALUATION_SCORE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_SCORE) | `"evaluation.score"` |
| [`EVALUATION_LABEL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_LABEL) | `"evaluation.label"` |
| [`EVALUATION_EXPLANATION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_EXPLANATION) | `"evaluation.explanation"` |
| [`EVALUATION_ANNOTATOR_KIND`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_ANNOTATOR_KIND) | `"evaluation.annotator_kind"` |
| [`EVALUATION_IDENTIFIER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_IDENTIFIER) | `"evaluation.identifier"` |
| [`EVALUATION_METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EVALUATION_METADATA) | `"evaluation.metadata"` |
| [`INPUT_VALUE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#INPUT_VALUE) | `"input.value"`The input to any span |
| [`INPUT_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#INPUT_MIME_TYPE) | `"input.mime_type"` |
| [`INPUT_IMAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#INPUT_IMAGES) | `"input.images"`Images passed as input to a span of any kind. Flattened with an index, for example `input.images.0.image.url`. |
| [`INPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#INPUT_DOCUMENTS) | `"input.documents"`Documents (e.g. PDFs) passed as input to a span of any kind. Flattened with an index, for example `input.documents.0.document.url`. |
| [`OUTPUT_VALUE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OUTPUT_VALUE) | `"output.value"`The output of any span |
| [`OUTPUT_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OUTPUT_MIME_TYPE) | `"output.mime_type"` |
| [`OUTPUT_IMAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OUTPUT_IMAGES) | `"output.images"`Images produced as output by a span of any kind. Flattened with an index, for example `output.images.0.image.url`. |
| [`OUTPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#OUTPUT_DOCUMENTS) | `"output.documents"`Documents (e.g. PDFs) produced as output by a span of any kind. Flattened with an index, for example `output.documents.0.document.url`. |
| [`LLM_INPUT_MESSAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_INPUT_MESSAGES) | `"llm.input_messages"`The messages sent to the LLM for completions Typically seen in OpenAI chat completions> **See also**
>
> <https://beta.openai.com/docs/api-reference/completions/create> |
| [`LLM_PROMPTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_PROMPTS) | `"llm.prompts"`The prompts sent to the LLM for completions Typically seen in OpenAI legacy completions> **See also**
>
> <https://beta.openai.com/docs/api-reference/completions/create> |
| [`LLM_INVOCATION_PARAMETERS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_INVOCATION_PARAMETERS) | `"llm.invocation_parameters"`The JSON representation of the parameters passed to the LLM |
| [`LLM_OUTPUT_MESSAGES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_OUTPUT_MESSAGES) | `"llm.output_messages"`The messages received from the LLM for completions Typically seen in OpenAI chat completions> **See also**
>
> <https://platform.openai.com/docs/api-reference/chat/object#choices-message> |
| [`LLM_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_MODEL_NAME) | `"llm.model_name"`The name of the LLM model |
| [`LLM_REQUEST_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_REQUEST_MODEL_NAME) | `"llm.request.model_name"`The model requested by the caller, as sent in the request. May differ from llm.response.model_name when the provider routes the request to a different model (e.g. classifier-triggered fallback). |
| [`LLM_RESPONSE_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_RESPONSE_MODEL_NAME) | `"llm.response.model_name"`The model that actually generated the response, as reported by the provider. May differ from llm.request.model_name when the provider routes the request to a different model (e.g. classifier-triggered fallback). |
| [`LLM_PROVIDER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_PROVIDER) | `"llm.provider"`The provider of the inferences. E.g. the cloud provider |
| [`LLM_SYSTEM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_SYSTEM) | `"llm.system"`The AI product as identified by the client or server |
| [`DECISION_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_MODEL_NAME) | `"decision.model_name"`The name of the decision model. Mirrors LLM_MODEL_NAME for DECISION spans. A decision model takes state plus typed questions and returns a typed, probabilistic answer per question instead of generated text.> **See also**
>
> <https://github.com/Arize-ai/openinference/blob/main/spec/spans/decision.md> |
| [`DECISION_REQUEST_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_REQUEST_MODEL_NAME) | `"decision.request.model_name"`The decision model requested by the caller, as sent in the request. May differ from decision.response.model_name when the provider resolves an alias (e.g. jev-latest) or routes the request to a different model. |
| [`DECISION_RESPONSE_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_RESPONSE_MODEL_NAME) | `"decision.response.model_name"`The decision model that actually produced the response, as reported by the provider. May differ from decision.request.model_name. |
| [`DECISION_PROVIDER`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_PROVIDER) | `"decision.provider"`The hosting provider of the decision model: who runs the hardware that answered (e.g. “typesafe” when calling TypeSafe directly, or the cloud or self-hosting provider for a model served elsewhere). Distinct from DECISION_SYSTEM, which names the API shape. Well-known values are [`DecisionProvider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionProvider), which alias the matching [`LLMProvider`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMProvider) values. |
| [`DECISION_SYSTEM`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_SYSTEM) | `"decision.system"`The decision API ecosystem the call conforms to, i.e. which request and response shape the client speaks, as identified by the client or server: “typesafe” for the TypeSafe System One / Jev API, “openai” for the OpenAI Decisions API. A self-hosted vLLM server answering the Jev-compatible /v1/systemone shape is still “typesafe”. Distinct from DECISION_PROVIDER, which says who hosts the model. Well-known values are [`DecisionSystem`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DecisionSystem), which alias the matching [`LLMSystem`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLMSystem) values. |
| [`DECISION_TOKEN_COUNT_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_TOKEN_COUNT_INPUT) | `"decision.token_count.input"`The number of input tokens consumed by a decision model call: the state, questions, and candidate options. Maps to usage.input_tokens in TypeSafe System One responses. |
| [`DECISION_TOKEN_COUNT_OUTPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DECISION_TOKEN_COUNT_OUTPUT) | `"decision.token_count.output"`The number of output tokens produced by a decision model call. Decision models emit typed answers rather than text, so this is typically small. Maps to usage.output_tokens in TypeSafe System One responses. |
| [`LLM_TOKEN_COUNT_COMPLETION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_COMPLETION) | `"llm.token_count.completion"`Token count for the completion by the llm (in tokens) |
| [`LLM_TOKEN_COUNT_COMPLETION_DETAILS_REASONING`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_COMPLETION_DETAILS_REASONING) | `"llm.token_count.completion_details.reasoning"`Token count for the reasoning steps in the completion (in tokens) |
| [`LLM_TOKEN_COUNT_COMPLETION_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_COMPLETION_DETAILS_AUDIO) | `"llm.token_count.completion_details.audio"`Token count for audio input generated by the model (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_PROMPT) | `"llm.token_count.prompt"`Token count for the prompt to the llm (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_WRITE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_WRITE) | `"llm.token_count.prompt_details.cache_write"`Token count for the tokens written to cache (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_READ`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_READ) | `"llm.token_count.prompt_details.cache_read"`Token count for the tokens retrieved from cache (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_PROMPT_DETAILS_CACHE_INPUT) | `"llm.token_count.prompt_details.cache_input"`Token count for the input tokens in the prompt that were cached (in tokens) |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_PROMPT_DETAILS_AUDIO) | `"llm.token_count.prompt_details.audio"`Token count for audio input presented in the prompt (in tokens) |
| [`LLM_TOKEN_COUNT_TOTAL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_TOTAL) | `"llm.token_count.total"`Token count for the entire transaction with the llm (in tokens) |
| [`LLM_FINISH_REASON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_FINISH_REASON) | `"llm.finish_reason"`The reason the model stopped generating tokens, e.g. `"stop"` or `"length"`. |
| [`LLM_TOKEN_COUNT_PROMPT_DETAILS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_PROMPT_DETAILS) | `"llm.token_count.prompt_details"`Key prefix for additional prompt token count details. Each detail should be a separate attribute with this prefix, e.g. llm.token_count.prompt_details.reasoning, llm.token_count.prompt_details.audio. All values should be in tokens (integer count of tokens). |
| [`LLM_TOKEN_COUNT_COMPLETION_DETAILS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOKEN_COUNT_COMPLETION_DETAILS) | `"llm.token_count.completion_details"`Key prefix for additional completion token count details. Each detail should be a separate attribute with this prefix, e.g. llm.token_count.completion_details.reasoning, llm.token_count.completion_details.audio. All values should be in tokens (integer count of tokens). |
| [`LLM_COST`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST) | `"llm.cost"`Key prefix for cost information. When these keys are transformed into a JSON-like structure, it would look like: { “prompt”: 0.0021, # Cost in USD “completion”: 0.0045, # Cost in USD “total”: 0.0066, # Cost in USD “completion_details”: { “output”: 0.0009, # Cost in USD “reasoning”: 0.0024, # Cost in USD (e.g., 80 tokens \* $0.03/1K tokens) “audio”: 0.0012 # Cost in USD (e.g., 40 tokens \* $0.03/1K tokens) }, “prompt_details”: { “input”: 0.0003, # Cost in USD “cache_write”: 0.0006, # Cost in USD (e.g., 20 tokens \* $0.03/1K tokens) “cache_read”: 0.0003, # Cost in USD (e.g., 10 tokens \* $0.03/1K tokens) “cache_input”: 0.0006, # Cost in USD (e.g., 20 tokens \* $0.03/1K tokens) “audio”: 0.0003 # Cost in USD (e.g., 10 tokens \* $0.03/1K tokens) } } Note: This is a key prefix - individual attributes are stored as separate span attributes with this prefix, e.g. llm.cost.prompt, llm.cost.completion_details.reasoning, etc. The JSON structure shown above represents how these separate attributes can be conceptually organized. All monetary values are in USD with floating point precision. |
| [`LLM_COST_PROMPT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_PROMPT) | `"llm.cost.prompt"`Cost of the prompt tokens in USD |
| [`LLM_COST_COMPLETION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_COMPLETION) | `"llm.cost.completion"`Cost of the completion tokens in USD |
| [`LLM_COST_TOTAL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_TOTAL) | `"llm.cost.total"`Total cost of the LLM call in USD (prompt + completion) |
| [`LLM_COST_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_INPUT) | `"llm.cost.prompt_details.input"`Total cost of input tokens in USD. This represents the cost of tokens that were used as input to the model, which may be different from the prompt cost if there are additional processing steps. |
| [`LLM_COST_OUTPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_OUTPUT) | `"llm.cost.completion_details.output"`Total cost of output tokens in USD. This represents the cost of tokens that were generated as output by the model, which may be different from the completion cost if there are additional processing steps. |
| [`LLM_COST_COMPLETION_DETAILS_REASONING`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_COMPLETION_DETAILS_REASONING) | `"llm.cost.completion_details.reasoning"`Cost of reasoning steps in the completion in USD |
| [`LLM_COST_COMPLETION_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_COMPLETION_DETAILS_AUDIO) | `"llm.cost.completion_details.audio"`Cost of audio tokens in the completion in USD |
| [`LLM_COST_PROMPT_DETAILS_CACHE_WRITE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_PROMPT_DETAILS_CACHE_WRITE) | `"llm.cost.prompt_details.cache_write"`Cost of prompt tokens written to cache in USD |
| [`LLM_COST_PROMPT_DETAILS_CACHE_READ`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_PROMPT_DETAILS_CACHE_READ) | `"llm.cost.prompt_details.cache_read"`Cost of prompt tokens read from cache in USD |
| [`LLM_COST_PROMPT_DETAILS_CACHE_INPUT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_PROMPT_DETAILS_CACHE_INPUT) | `"llm.cost.prompt_details.cache_input"`Cost of input tokens in the prompt that were cached in USD |
| [`LLM_COST_PROMPT_DETAILS_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_COST_PROMPT_DETAILS_AUDIO) | `"llm.cost.prompt_details.audio"`Cost of audio tokens in the prompt in USD |
| [`MESSAGE_ROLE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_ROLE) | `"message.role"`The role that the LLM assumes the message is from during the LLM invocation |
| [`MESSAGE_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_NAME) | `"message.name"`The name of the message. This is only used for role ‘function’ where the name of the function is captured in the name field and the parameters are captured in the content. |
| [`MESSAGE_TOOL_CALLS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_TOOL_CALLS) | `"message.tool_calls"`The tool calls generated by the model, such as function calls. |
| [`MESSAGE_TOOL_CALL_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_TOOL_CALL_ID) | `"message.tool_call_id"`The id of the tool call on a “tool” role message |
| [`TOOL_CALL_FUNCTION_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_CALL_FUNCTION_NAME) | `"tool_call.function.name"`tool_call.function.name |
| [`TOOL_CALL_FUNCTION_ARGUMENTS_JSON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_CALL_FUNCTION_ARGUMENTS_JSON) | `"tool_call.function.arguments"`tool_call.function.argument (JSON string) |
| [`TOOL_CALL_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_CALL_ID) | `"tool_call.id"`The id of the tool call |
| [`TOOL_CALL_REASONING_SIGNATURE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_CALL_REASONING_SIGNATURE) | `"tool_call.reasoning_signature"`Opaque vendor-issued reasoning echo token attached to a tool call. Maps to Gemini thoughtSignature when it is attached to a functionCall part |
| [`MESSAGE_FUNCTION_CALL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_FUNCTION_CALL_NAME) | `"message.function_call_name"`The LLM function call function name |
| [`MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_FUNCTION_CALL_ARGUMENTS_JSON) | `"message.function_call_arguments_json"`The LLM function call function arguments in a json string |
| [`MESSAGE_CONTENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT) | `"message.content"`The content of the message sent to the LLM |
| [`MESSAGE_CONTENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENTS) | `"message.contents"`The array of contents for the message sent to the LLM. Each element of the array is an `message_content` object. |
| [`MESSAGE_CONTENT_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_TYPE) | `"message_content.type"`The type of content sent to the LLM, such as “text”, “image”, “audio”, “video”, “document”, “reasoning”, or “tool_use” |
| [`MESSAGE_CONTENT_TEXT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_TEXT) | `"message_content.text"`The text content of the message sent to the LLM |
| [`MESSAGE_CONTENT_IMAGE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_IMAGE) | `"message_content.image"`The image content of the message sent to the LLM |
| [`MESSAGE_CONTENT_AUDIO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_AUDIO) | `"message_content.audio"`The audio content of the message sent to the LLM |
| [`MESSAGE_CONTENT_VIDEO`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_VIDEO) | `"message_content.video"`The video content of the message sent to the LLM |
| [`MESSAGE_CONTENT_DOCUMENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_DOCUMENT) | `"message_content.document"`The document content (e.g. a PDF, an uploaded file, or inline text) of the message sent to the LLM |
| [`MESSAGE_CONTENT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_ID) | `"message_content.id"`Provider-assigned identifier for this message content item. For OpenAI Responses reasoning items, this maps to ResponseReasoningItem.id and should be preserved for stateless replay |
| [`MESSAGE_CONTENT_SIGNATURE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_SIGNATURE) | `"message_content.signature"`Opaque vendor-issued signature captured verbatim. Maps to provider signature fields and to Gemini thoughtSignature fields when the signature is attached to a non-tool content part |
| [`MESSAGE_CONTENT_DATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_DATA) | `"message_content.data"`Opaque vendor-issued data captured verbatim. Maps to Anthropic redacted_thinking.data |
| [`MESSAGE_CONTENT_ENCRYPTED_CONTENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#MESSAGE_CONTENT_ENCRYPTED_CONTENT) | `"message_content.encrypted_content"`OpenAI encrypted_content captured verbatim |
| [`IMAGE_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#IMAGE_URL) | `"image.url"`The http or base64 link to the image |
| [`VIDEO_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#VIDEO_URL) | `"video.url"`The URL, object-store URI, or base64 data URI of a video. |
| [`DOCUMENT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_ID) | `"document.id"` |
| [`DOCUMENT_CONTENT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_CONTENT) | `"document.content"` |
| [`DOCUMENT_SCORE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_SCORE) | `"document.score"` |
| [`DOCUMENT_METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_METADATA) | `"document.metadata"` |
| [`DOCUMENT_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_URL) | `"document.url"`The URL, object-store URI, or base64 data URI of a file document such as a PDF. |
| [`DOCUMENT_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_MIME_TYPE) | `"document.mime_type"`The MIME type of a document, for example `application/pdf`. |
| [`DOCUMENT_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#DOCUMENT_NAME) | `"document.name"`The name supplied with a document, such as a file name or a title, for example `invoice.pdf`. Not a unique identifier. |
| [`EMBEDDING_TEXT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EMBEDDING_TEXT) | `"embedding.text"`The text that was embedded to create the vector |
| [`EMBEDDING_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EMBEDDING_MODEL_NAME) | `"embedding.model_name"`The name of the model that was used to create the vector |
| [`EMBEDDING_VECTOR`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EMBEDDING_VECTOR) | `"embedding.vector"`The embedding vector. Typically a high dimensional vector of floats or ints |
| [`EMBEDDING_EMBEDDINGS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#EMBEDDING_EMBEDDINGS) | `"embedding.embeddings"`The embedding list root |
| [`RETRIEVAL_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RETRIEVAL_DOCUMENTS) | `"retrieval.documents"`The retrieval documents list root |
| [`PROMPT_TEMPLATE_VARIABLES`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PROMPT_TEMPLATE_VARIABLES) | `"llm.prompt_template.variables"`The JSON representation of the variables used in the prompt template |
| [`PROMPT_TEMPLATE_TEMPLATE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PROMPT_TEMPLATE_TEMPLATE) | `"llm.prompt_template.template"`A prompt template |
| [`LLM_FUNCTION_CALL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_FUNCTION_CALL) | `"llm.function_call"`The JSON representation of a function call of an LLM |
| [`LLM_TOOLS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#LLM_TOOLS) | `"llm.tools"`List of tools that are advertised to the LLM to be able to call |
| [`TOOL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_NAME) | `"tool.name"`The name of a tool |
| [`TOOL_DESCRIPTION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_DESCRIPTION) | `"tool.description"`The description of a tool |
| [`TOOL_PARAMETERS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_PARAMETERS) | `"tool.parameters"`The parameters of the tool represented as a JSON string |
| [`TOOL_JSON_SCHEMA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_JSON_SCHEMA) | `"tool.json_schema"`The json schema of a tool input, It is RECOMMENDED that this be in the OpenAI tool calling format: <https://platform.openai.com/docs/assistants/tools> |
| [`TOOL_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TOOL_ID) | `"tool.id"`The identifier for the result of the tool call (corresponding to tool_call.id). |
| [`SESSION_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#SESSION_ID) | `"session.id"`The session id of a trace. Used to correlate spans in a single session. |
| [`USER_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#USER_ID) | `"user.id"`The user id of a trace. Used to correlate spans for a single user. |
| [`RERANKER_INPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RERANKER_INPUT_DOCUMENTS) | `"reranker.input_documents"`The documents used as input to the reranker |
| [`RERANKER_OUTPUT_DOCUMENTS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RERANKER_OUTPUT_DOCUMENTS) | `"reranker.output_documents"`The documents output by the reranker |
| [`RERANKER_QUERY`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RERANKER_QUERY) | `"reranker.query"`The query string for the reranker |
| [`RERANKER_MODEL_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RERANKER_MODEL_NAME) | `"reranker.model_name"`The model name for the reranker |
| [`RERANKER_TOP_K`](https://openinf.dev/js/openinference-semantic-conventions.html.md#RERANKER_TOP_K) | `"reranker.top_k"`The top k parameter for the reranker |
| [`METADATA`](https://openinf.dev/js/openinference-semantic-conventions.html.md#METADATA) | `"metadata"`Metadata for a span, used to store user-defined key-value pairs |
| [`PROMPT_TEMPLATE_VERSION`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PROMPT_TEMPLATE_VERSION) | `"llm.prompt_template.version"`A prompt template version |
| [`TAG_TAGS`](https://openinf.dev/js/openinference-semantic-conventions.html.md#TAG_TAGS) | `"tag.tags"`The tags associated with a span |
| [`AUDIO_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AUDIO_URL) | `"audio.url"`The url of an audio file |
| [`AUDIO_MIME_TYPE`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AUDIO_MIME_TYPE) | `"audio.mime_type"`The audio mime type |
| [`AUDIO_TRANSCRIPT`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AUDIO_TRANSCRIPT) | `"audio.transcript"`The audio transcript as text |
| [`PROMPT_VENDOR`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PROMPT_VENDOR) | `"prompt.vendor"`The vendor or origin of the prompt, e.g. a prompt library, a specialized service, etc. |
| [`PROMPT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PROMPT_ID) | `"prompt.id"`A vendor-specific id used to locate the prompt |
| [`PROMPT_URL`](https://openinf.dev/js/openinference-semantic-conventions.html.md#PROMPT_URL) | `"prompt.url"`A vendor-specific URL used to locate the prompt |
| [`AGENT_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#AGENT_NAME) | `"agent.name"`The name of the agent. Agents that perform the same functions should have the same name. |
| [`GRAPH_NODE_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#GRAPH_NODE_ID) | `"graph.node.id"`The id of the node in the execution graph. This along with graph.node.parent_id are used to visualize the execution graph. |
| [`GRAPH_NODE_NAME`](https://openinf.dev/js/openinference-semantic-conventions.html.md#GRAPH_NODE_NAME) | `"graph.node.name"`The name of the node in the execution graph. Use this to present a human readable name for the node. Optional |
| [`GRAPH_NODE_PARENT_ID`](https://openinf.dev/js/openinference-semantic-conventions.html.md#GRAPH_NODE_PARENT_ID) | `"graph.node.parent_id"`This references the id of the parent node. Leaving this unset or set as empty string implies that the current span is the root node. |
