---
title: "Get Context Usage"
method: GET
path: "/api/chats/{chat_id}/context-usage"
tags: ["chats"]
---

# Get Context Usage

`GET /api/chats/{chat_id}/context-usage`

## Request

- Base URL: `https://app.thinqos.com` (the document declares no server; this is the origin it was published from)
- URL: `https://app.thinqos.com/api/chats/{chat_id}/context-usage`
- Auth: none declared

The logged-in viewer's OWN context-window usage for this chat (TOS-761).

Per-viewer: you never see another participant's context window. 404 if the
caller is not an active participant.

## Path parameters

- `chat_id` string, uuid, required

## Headers

- `x-impersonate-identity-id` string, nullable

## Response `200`

Successful Response

- ContextUsageResponse: The logged-in viewer's OWN context-window usage for a chat (TOS-761). Strictly scoped to the caller's identity: every identity maintains its own context window (humans included, via augmented/agent mode), keyed by ``(chat, identity)``, so you never see another participant's. ``usage_percent`` is honest *token fill*: it uses the post-boundary high-water chat model-run prompt tokens when available, otherwise the viewer's active rolling summary + bounded raw window (``used_tokens``), measured against the resolved model input ceiling (``context_window``). ``floor_tokens`` is the chat-independent fixed prompt cost and ``history_tokens`` is the chat's summary plus bounded raw-window contribution. The token figures are a UI gauge, not billing numbers.
  - `context_window` integer, required
  - `used_tokens` integer, required
  - `usage_percent` integer, required
  - `floor_tokens` integer
  - `history_tokens` integer

## Other responses

- `422`: Validation Error

---

[API](https://skmtc.dev/thinqos/apis/thinqos-api.md) · [All operations](https://skmtc.dev/thinqos/apis/thinqos-api/llms.txt) · [OpenAPI document](https://skmtc.dev/thinqos/apis/thinqos-api/revisions/5387b4ded614?raw)