> ## Documentation Index > Fetch the complete documentation index at: https://docs.polymarket.com/llms.txt > Use this file to discover all available pages before exploring further. # 获取交易所统计数据 > Get aggregate statistics for all pUSD-quoted Polymarket perpetual markets. The time window is half-open: trades at `start_timestamp` are included and trades at `end_timestamp` are excluded. The maximum window is 31 days. Open interest is the one-sided USD notional from the latest complete sample before `end_timestamp`; its value and timestamp are null when no complete sample is available. Fees are gross positive maker and taker trading fees. Rebates, incentives, and referral payments are excluded. ## OpenAPI ````yaml api-spec/perps-openapi.json GET /v1/info/exchange-stats openapi: 3.0.3 info: title: Polymarket Perps HTTP API version: 1.0.0 description: HTTP API for Polymarket perpetual trading system. license: name: Apache 2.0 url: https://www.apache.org/licenses/LICENSE-2.0.html servers: - url: https://api.perpetuals.polymarket.com description: Production Perps HTTP API security: [] paths: /v1/info/exchange-stats: get: summary: Get Exchange Statistics description: > Get aggregate statistics for all pUSD-quoted Polymarket perpetual markets. The time window is half-open: trades at `start_timestamp` are included and trades at `end_timestamp` are excluded. The maximum window is 31 days. Open interest is the one-sided USD notional from the latest complete sample before `end_timestamp`; its value and timestamp are null when no complete sample is available. Fees are gross positive maker and taker trading fees. Rebates, incentives, and referral payments are excluded. operationId: getExchangeStatistics parameters: - name: start_timestamp in: query required: true schema: $ref: '#/components/schemas/start_timestamp' - name: end_timestamp in: query required: true schema: $ref: '#/components/schemas/end_timestamp' responses: '200': description: Exchange statistics response. content: application/json: schema: $ref: '#/components/schemas/ExchangeStatistics' '400': $ref: '#/components/responses/Error400Response' '408': $ref: '#/components/responses/Error408Response' '413': $ref: '#/components/responses/Error413Response' '429': $ref: '#/components/responses/Error429Response' '500': $ref: '#/components/responses/Error500Response' '503': $ref: '#/components/responses/Error503Response' security: [] components: schemas: start_timestamp: type: integer description: Start timestamp in milliseconds example: 1767225600000 end_timestamp: type: integer description: End timestamp in milliseconds example: 1767229200000 ExchangeStatistics: type: object required: - start_timestamp - end_timestamp - volume - open_interest - open_interest_timestamp - fees properties: start_timestamp: $ref: '#/components/schemas/start_timestamp' end_timestamp: $ref: '#/components/schemas/end_timestamp' volume: $ref: '#/components/schemas/exchange_volume' open_interest: $ref: '#/components/schemas/exchange_open_interest' open_interest_timestamp: $ref: '#/components/schemas/open_interest_timestamp' fees: $ref: '#/components/schemas/exchange_fees' exchange_volume: type: string description: >- Matched trading volume in USD notional for the requested window, as an unformatted numeric string example: '10164976.25' exchange_open_interest: type: string nullable: true description: >- One-sided open interest in USD notional at the latest complete sample before the requested window end, as an unformatted numeric string. Null when no complete sample is available. example: '9238345.12' open_interest_timestamp: type: integer nullable: true description: >- Timestamp of the open-interest sample in Unix milliseconds. Null when no complete sample is available. example: 1767225570000 exchange_fees: type: string description: >- Gross positive maker and taker trading fees in USD for the requested window, as an unformatted numeric string. Rebates, incentives, and referral payments are excluded. example: '1245.67' Error400: title: Error400 type: object required: - status - error properties: status: type: string enum: - err error: $ref: '#/components/schemas/error' Error408: title: Error408 type: object required: - status - error properties: status: type: string enum: - err error: $ref: '#/components/schemas/error' Error413: title: Error413 type: object required: - status - error properties: status: type: string enum: - err error: $ref: '#/components/schemas/error' Error429: title: Error429 type: object required: - status - error properties: status: type: string enum: - err error: $ref: '#/components/schemas/error' Error500: title: Error500 type: object required: - status - error properties: status: type: string enum: - err error: $ref: '#/components/schemas/error' Error503: title: Error503 type: object required: - status - error properties: status: type: string enum: - err error: $ref: '#/components/schemas/error' error: type: string description: >- Error identifier. For domain rejections and transport errors (`401`/`404`/`429`/`500`) this is a stable, machine-readable snake_case identifier that is part of the API contract and safe to branch on, e.g. `insufficient_margin`, `insufficient_balance`, `order_not_found`, `reduce_only_invalid`, `price_outside_bounds`, `position_not_found`, `position_exists`, `open_orders_exist`, `invalid_margin_mode`, `invalid_margin_amount`, `margin_below_required_initial`, `account_liquidating`, `unauthorized`, `not_found`. For `400` it is a human-readable validation detail whose wording may change. See the Error handling guide for the domain identifiers. (Post-only / Fill-or-Kill outcomes are order statuses such as `post_only_rejected`, not rejections.) example: insufficient_margin responses: Error400Response: description: | Bad request — the request was malformed or failed validation (bad query parameters, unparseable body, invalid signature, or a domain pre-check). The `error` field is a human-readable validation detail. content: application/json: schema: $ref: '#/components/schemas/Error400' Error408Response: description: | Request Timeout — the request body was not delivered within the gateway's read deadline. Protects against slow uploads holding connections open; send the complete body promptly and retry. `error` is `request_body_timeout`. content: application/json: schema: $ref: '#/components/schemas/Error408' Error413Response: description: | Content Too Large — the request body exceeds the gateway's size cap, judged on the declared `Content-Length` or the actual stream. Split oversized batches into smaller requests. `error` is `payload_too_large`. content: application/json: schema: $ref: '#/components/schemas/Error413' Error429Response: description: > Too Many Requests. `error` distinguishes the limit that was hit: `ip_rate_limited` (per-IP token bucket), `action_rate_limited` (per-account action rate), or `open_orders_limit` (resting open-order cap). headers: Retry-After: description: > Whole seconds to wait before retrying. Present only on token-bucket rate-limit rejections (`ip_rate_limited` and `action_rate_limited`); a conservative estimate of when enough capacity will have refilled to admit the request. Absent on `open_orders_limit`, which is a capacity limit, not a rate limit — waiting does not free order slots; cancel resting orders or wait for fills instead. schema: type: integer example: 2 content: application/json: schema: $ref: '#/components/schemas/Error429' Error500Response: description: | Internal server error. `error` is `internal_error`. content: application/json: schema: $ref: '#/components/schemas/Error500' Error503Response: description: | Service unavailable — safe to retry with backoff, honoring `Retry-After` when present. `error` is `service_unavailable`. Most causes are a request deliberately shed before it reached the engine, so the request definitely did not execute: the replica's pod-level in-flight cap (shed before dispatch, with `Retry-After`), a database read shed as backpressure under overload (by the local read gate or by the database server's concurrency cap — the query never executed), a required internal cache that is stale or not yet primed and would return wrong data, or a full bounded deferred-write queue. On a subset of order/state-mutating routes, this status is also returned when the engine round-trip itself times out with zero responses observed. That is retryable engine backpressure, but — unlike the sheds above — it does not prove non-execution: the command may already have been admitted, with its response simply not reaching the gateway before the deadline. This status is only used for that case on routes that are safe to retry regardless (idempotent state-setting, or an engine-side replay guard); routes where a bare retry cannot safely resolve the ambiguity keep the outcome a non-retryable 500 instead. headers: Retry-After: description: | Whole seconds to wait before retrying. Present on pre-dispatch in-flight-cap sheds — a transient overload signal, so the hint is short and constant. schema: type: integer example: 1 content: application/json: schema: $ref: '#/components/schemas/Error503' ````