> ## Documentation Index > Fetch the complete documentation index at: https://vastai-80aa3a82-host-2752-sla-api-documentation.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # sla backtest > Replays a machine's historical compute earnings and uptime through SLA settlement math to preview SLA charges and credits before listing with SLA. The maximum date range is 366 days (inclusive); the default lookback is 90 days when custom dates are omitted. Requires `machine_read` permission. See [SLA Earnings Backtester](/host/sla-backtester) for usage guide. ## OpenAPI ````yaml /api-reference/openapi.yaml get /api/v0/host/sla_backtest openapi: 3.1.0 info: title: Vast.ai API description: >- Vast.ai REST API for managing GPU cloud instances, machine operations, and AI/ML workflows. ## AI Agent Quick-Start Install the CLI skill for your agent (Claude Code, Cursor, Windsurf, etc.): npx skills add vast-ai/vast-cli CLI reference: https://raw.githubusercontent.com/vast-ai/vast-cli/master/vastai/SKILL.md SDK reference: https://raw.githubusercontent.com/vast-ai/vast-cli/master/vastai_sdk/SKILL.md ## Auth All endpoints require `Authorization: Bearer $VAST_API_KEY`. Get your key at: https://cloud.vast.ai/manage-keys/ ## Key Quirks - `gpu_ram` in CLI = GB; in REST API = MB (CLI auto-converts) - SSH keys must be registered BEFORE creating an instance (VM: no recovery; Docker: can add post-create) - `onstart` field is limited to 4048 characters -- gzip+base64 for longer scripts - `POST /api/v0/asks/{id}/` (create instance) returns `new_contract` as the instance ID, not `id` - Poll trap: if `actual_status` becomes `exited`, `unknown`, or `offline` it will never reach `running` -- destroy and retry version: 1.0.0 contact: name: Vast.ai Support url: https://discord.gg/vast license: name: Vast.ai Terms of Service url: https://vast.ai/terms/ servers: - url: https://console.vast.ai description: Production server security: - BearerAuth: [] paths: /api/v0/host/sla_backtest: get: tags: - Host summary: sla backtest description: >- Replays a machine's historical compute earnings and uptime through SLA settlement math to preview SLA charges and credits before listing with SLA. The maximum date range is 366 days (inclusive); the default lookback is 90 days when custom dates are omitted. Requires `machine_read` permission. See [SLA Earnings Backtester](/host/sla-backtester) for usage guide. operationId: slaBacktest parameters: - name: machine_id in: query required: true schema: type: integer description: ID of your machine. - name: r in: query required: true schema: type: number format: float minimum: 0 exclusiveMinimum: true maximum: 0.9999 description: >- Reliability level to simulate (greater than 0 and at most 0.9999). Used as both claim and client target in the simulation. Exactly 1.0 is rejected. - name: p in: query schema: type: number format: float minimum: 0 exclusiveMinimum: true exclusiveMaximum: true maximum: 1 default: 0.6667 description: >- Host confidence (greater than 0 and less than 1). Maps to sla_sigma_x when listing. Default 0.6667 (normal). - name: start_date in: query schema: type: integer description: >- Range start as Unix epoch seconds. Default is 90 days before end_date. - name: end_date in: query schema: type: integer description: Range end as Unix epoch seconds. Default is now. - name: reliability in: query schema: type: number format: float minimum: 0 exclusiveMinimum: true maximum: 1 description: >- Optional flat uptime override (greater than 0 and at most 1) for all replayed days. Exactly 0 is rejected. - name: beta in: query schema: type: number format: float default: 0.5 description: > Timescale mix used in settlement simulation. Not user-configurable for host API keys — the API derives weighting from `r` (recent vs longer-term uptime). Omit this parameter. responses: '200': description: Successful response content: application/json: schema: type: object properties: success: type: boolean machine_id: type: integer has_data: type: boolean description: false when no days in the range had earnings to replay. start_day: type: integer description: UTC day index (days since 1970-01-01). end_day: type: integer description: UTC day index (days since 1970-01-01). covered_days: type: integer skipped_days: type: integer reliability: type: number format: float fleet_reliability: type: number format: float description: >- Average reliability across your machines (including this one). Context only. params: type: object properties: r: type: number format: float p: type: number format: float sigma_x: type: number format: float reliability: type: - number - 'null' format: float beta: type: number format: float daily_results: type: array items: type: object properties: day: type: integer gpu: type: number format: float storage: type: number format: float bandwidth: type: number format: float uptime: type: number format: float sla: type: number format: float historical_sla: type: number format: float summary: type: object properties: total_gpu: type: number format: float total_storage: type: number format: float total_bandwidth: type: number format: float total_sla: type: number format: float total_earnings: type: number format: float total_historical_sla: type: number format: float worst_day_loss: type: number format: float reserve: type: number format: float '400': description: Bad Request content: application/json: schema: type: object properties: error: type: string enum: - invalid_range - range_too_large - bad_request - invalid_params msg: type: string '403': description: Forbidden — not machine owner content: application/json: schema: type: object properties: error: type: string example: not_owner '404': description: Machine not found content: application/json: schema: type: object properties: error: type: string example: machine_not_found '429': description: Too Many Requests — rate limit exceeded security: - BearerAuth: [] components: securitySchemes: BearerAuth: type: http scheme: bearer description: API key must be provided in the Authorization header ````