---
title: "Get Chat Usage"
method: GET
path: "/agent/chat/{chat_id}/usage"
tags: ["accepts-api-key"]
---

# Get Chat Usage

`GET /agent/chat/{chat_id}/usage`

## Request

- Base URL: `https://api.tellers.ai` (the document declares no server; this is the origin it was published from)
- URL: `https://api.tellers.ai/agent/chat/{chat_id}/usage`
- Auth: none declared

Return chat-token and per-tool token usage for a chat, and its sub-chats' total.

## Path parameters

- `chat_id` string, uuid, required

## Headers

- `x-api-key` string, nullable
- `authorization` string, nullable

## Response `200`

Successful Response

- ChatUsageResponse
  - `chat_id` string, required
  - `chat_tokens_used` number, required
  - `tools_tokens_used` number, required
  - `sub_chats_tokens_used` number: What this chat's sub-chats spent, agent and tools together. Already counted in `total_tokens_used`; `chat_tokens_used`, `tools` and `breakdown` cover this chat only.
  - `sub_chat_count` integer: How many sub-chats this chat has started.
  - `total_tokens_used` number, required: This chat's spend plus its sub-chats'.
  - `tools` ChatToolUsage[], required
    - `tool_name` string, required
    - `calls` integer, required
    - `tokens_used` number, required
  - `breakdown` ChatUsageBreakdown: Where a chat's tokens went, as raw model tokens rather than charges. The first four counts come from OpenAI; the ``*_tokens`` counts below them are our own tiktoken measurements of the parts we compose the prompt from, which is the only way to attribute a cost to tool schemas, project state or replayed history — the API reports the input as one number. ``complete`` is false when the chat contains calls made before these counts were recorded. The numbers are then a floor for the chat, not its total.
    - `llm_calls` integer
    - `input_tokens_uncached` integer
    - `input_tokens_cached` integer
    - `cache_write_tokens` integer
    - `output_tokens` integer
    - `reasoning_tokens` integer
    - `system_prompt_tokens` integer
    - `assistant_prompt_tokens` integer
    - `project_state_tokens` integer
    - `conversation_history_tokens` integer
    - `user_message_tokens` integer
    - `tool_result_tokens` integer
    - `tool_schema_tokens` integer
    - `timeline_mutations` integer
    - `llm_calls_per_timeline_mutation` number, nullable
    - `complete` boolean

## Other responses

- `422`: Validation Error

---

[API](https://skmtc.dev/tellers/apis/tellers-api.md) - [All operations](https://skmtc.dev/tellers/apis/tellers-api/llms.txt) - [OpenAPI document](https://skmtc.dev/tellers/apis/tellers-api/revisions/fdaa0fa44a26?raw)