> ## Documentation Index > Fetch the complete documentation index at: https://docs.lumisreach.com/llms.txt > Use this file to discover all available pages before exploring further. # Predict performance > Predict how a video, text post or image post will do before it goes out, with the models behind the dashboard's Predicted performance. Send the content, where it will be posted and, per platform, the account's follower count and past posts with their counts: the prediction ranks it against that account's own posts and scales the counts by what its usual post does. A video (`format: video`) is predicted for Instagram; a text post for X and LinkedIn; an image post for X, LinkedIn and Instagram (where the reel model reads its pictures). It runs in the background: this answers `202` with `status: processing`; poll `GET /v1/predictions/{id}` until it is `succeeded` (usually under a minute for a post, up to six for a video), then read `result.overall_score` (70 is a typical post for the account), the per-platform counts and what drove them. On the API & MCP plan and the Growth Marketing Manager hire each prediction is charged from the prepaid balance once it succeeds, at its estimated model cost plus 20% (`price_cents`, quoted in the response; a failed one is free); on Revenue Engine and Custom it is free; other plans get `403 plan_not_included`. Up to 5 run at once and 200 per day per workspace. ## OpenAPI ````yaml /openapi.json post /v1/predictions openapi: 3.1.0 info: title: LumisReach v1 API version: 1.0.0 description: >- Programmatic access to the LumisReach voice platform — calls, numbers, caller trust, and brands. contact: name: LumisReach email: founders@lumisreach.com servers: - url: https://api.lumisreach.com/api security: - bearerAuth: [] tags: - name: Contacts - name: Calls - name: Messages - name: Emails - name: Phone Numbers - name: Caller IDs - name: Enterprise Registration - name: Branded Calling - name: Dialer Calls - name: SMS - name: WebRTC - name: Reputation - name: Usage - name: Account - name: Billing - name: Documents - name: Lincoln AI - name: Optimized posts - name: Predictions - name: Quick Agent Actions - name: Sub-entities - name: Video schedules - name: Videos paths: /v1/predictions: post: tags: - Predictions summary: Predict performance description: >- Predict how a video, text post or image post will do before it goes out, with the models behind the dashboard's Predicted performance. Send the content, where it will be posted and, per platform, the account's follower count and past posts with their counts: the prediction ranks it against that account's own posts and scales the counts by what its usual post does. A video (`format: video`) is predicted for Instagram; a text post for X and LinkedIn; an image post for X, LinkedIn and Instagram (where the reel model reads its pictures). It runs in the background: this answers `202` with `status: processing`; poll `GET /v1/predictions/{id}` until it is `succeeded` (usually under a minute for a post, up to six for a video), then read `result.overall_score` (70 is a typical post for the account), the per-platform counts and what drove them. On the API & MCP plan and the Growth Marketing Manager hire each prediction is charged from the prepaid balance once it succeeds, at its estimated model cost plus 20% (`price_cents`, quoted in the response; a failed one is free); on Revenue Engine and Custom it is free; other plans get `403 plan_not_included`. Up to 5 run at once and 200 per day per workspace. operationId: v1_predictions_post parameters: - name: Idempotency-Key in: header required: false description: >- A UUID. When present, deduplicates repeat submissions. See /api-reference/idempotency. schema: type: string requestBody: required: true content: application/json: schema: $schema: http://json-schema.org/draft-07/schema# type: object properties: format: type: string enum: - video - text - image description: >- `video` (a reel), `text` (text alone) or `image` (text with one or more pictures) platforms: minItems: 1 maxItems: 6 type: array items: type: string enum: - tiktok - instagram - facebook - x - linkedin - youtube description: >- Where it will be posted. `video`: `instagram`. `text`: `x` and/or `linkedin`. `image`: any of `x`, `linkedin` and `instagram`. text: default: '' description: >- The caption (video) or post text (text, image), hashtags included, as it will be posted type: string maxLength: 10000 video_url: description: '`video`: a public https URL of the finished MP4' type: string format: uri image_urls: description: >- `image`: public https URLs of the pictures, in order. X and LinkedIn read the first four, Instagram the first ten; all are counted. maxItems: 20 type: array items: type: string format: uri post_at: description: When it will go out (ISO 8601); now when omitted type: string format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z|([+-](?:[01]\d|2[0-3]):[0-5]\d)))$ accounts: default: {} description: >- Per platform: the account's follower count and posting history. Omit for a new account. type: object propertyNames: type: string enum: - tiktok - instagram - facebook - x - linkedin - youtube additionalProperties: type: object properties: followers: description: >- The account's follower count, to size an account with little history type: integer minimum: 0 maximum: 9007199254740991 history: default: [] description: >- The account's own past posts on this platform, any order. Their counts set what a usual post does here (posts from the last 90 days that are at least two days old); for text and image posts the newest 20 are also read for how alike the new one is and how soon after the last it goes out. maxItems: 100 type: array items: type: object properties: text: default: '' description: Its caption or post text type: string maxLength: 10000 posted_at: type: string format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z|([+-](?:[01]\d|2[0-3]):[0-5]\d)))$ description: When it went out (ISO 8601) views: description: Plays, views or impressions type: integer minimum: 0 maximum: 9007199254740991 likes: type: integer minimum: 0 maximum: 9007199254740991 comments: type: integer minimum: 0 maximum: 9007199254740991 reposts: description: Reposts or shares type: integer minimum: 0 maximum: 9007199254740991 video_url: description: >- `video` only: a public URL of the reel. The newest five with one are watched next to the new video, which ranks it against your own work rather than the training accounts'. type: string format: uri required: - text - posted_at additionalProperties: false required: - history additionalProperties: false required: - format - platforms - text - accounts additionalProperties: false responses: '200': description: Success content: application/json: schema: $schema: http://json-schema.org/draft-07/schema# type: object properties: data: type: object properties: id: type: string format: uuid pattern: >- ^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$ status: type: string enum: - processing - succeeded - failed description: >- `processing` until it is done: usually under a minute for a post, up to six for a video format: type: string enum: - video - text - image platforms: type: array items: type: string enum: - tiktok - instagram - facebook - x - linkedin - youtube created_at: type: string format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ completed_at: anyOf: - type: string format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ - type: 'null' error: description: Why it failed; send it again once that is fixed type: - string - 'null' price_cents: anyOf: - type: integer minimum: -9007199254740991 maximum: 9007199254740991 - type: 'null' description: >- What it costs, in cents, quoted when it was asked for; null on a plan where predictions are free. Charged from the prepaid balance only once it succeeds. charged: type: boolean description: Whether `price_cents` has been charged result: anyOf: - type: object properties: overall_score: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- One 0-100 grade for how it is likely to do, read like a school grade: 70 is a typical post for the account, 85 a clearly better one, 55 a weaker one. overall_score_parts: type: array items: type: object properties: label: type: string chance: type: number description: >- The chance this input gives it of beating the usual post (0.5 is even) weight: type: number description: How much of it counts, 0 to 1 required: - label - chance - weight additionalProperties: false description: What the overall score is made of basis: type: string enum: - history - followers - benchmark description: >- What sets the usual post the counts scale: the account's `history`, its `followers`, or a `benchmark` for a new account quality: anyOf: - type: object properties: overall: type: number hook: type: object properties: score: type: number reason: type: string required: - score - reason additionalProperties: false script: type: object properties: score: type: number reason: type: string required: - score - reason additionalProperties: false video: type: object properties: score: type: number reason: type: string required: - score - reason additionalProperties: false source: type: string enum: - watch - review description: >- `watch`: an AI watched the video, or looked through an Instagram post's pictures; `review`: an AI read the post required: - overall - hook - script - video - source additionalProperties: false - type: 'null' description: >- 0-100 quality scores, 50 a routine post; null when the AI critique failed platforms: type: array items: type: object properties: platform: type: string enum: - tiktok - instagram - facebook - x - linkedin - youtube model_version: type: string percentile: type: number description: >- The share of the account's own posts this one should beat, 0 to 1 (0.5 is a typical post). A soft estimate: picking the better of two of an account's posts, the models are right about 6 times in 10. relative: type: object properties: engagement: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false likes: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false comments: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false views: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false reposts: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false conversion: description: >- `instagram`: engagement and buying-intent comments, the reel model's headline type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false required: - likes - comments additionalProperties: false description: >- Multiples of the account's usual post (1 is typical): p10, p50 and p90 predicted: type: object properties: views: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false likes: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false comments: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false engagement: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false buying_intent_comments: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false required: - views - likes - comments additionalProperties: false description: >- Counts, from `relative` and what a usual post does (see `basis`) drivers: type: array items: type: object properties: label: type: string impact: type: number description: >- How much it moved expected engagement: 0.25 is +25% detail: description: >- What the post has for it, e.g. "2 hashtags" type: - string - 'null' required: - label - impact - detail additionalProperties: false description: >- What in the content moved the prediction, largest first history_posts: type: integer minimum: -9007199254740991 maximum: 9007199254740991 description: >- How many of the account's past posts here were read required: - platform - model_version - percentile - relative - predicted - drivers - history_posts additionalProperties: false total: type: object properties: views: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false likes: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false comments: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false engagement: type: object properties: low: type: number expected: type: number high: type: number required: - low - expected - high additionalProperties: false required: - views - likes - comments additionalProperties: false likely_comments: type: array items: type: object properties: author: type: string text: type: string tone: type: string enum: - positive - question - neutral - critical required: - author - text - tone additionalProperties: false description: >- Simulated: the comments viewers are likely to leave tips: type: array items: type: string description: Concrete changes that would raise reach required: - overall_score - overall_score_parts - basis - quality - platforms - total - likely_comments - tips additionalProperties: false - type: 'null' description: Set once `status` is `succeeded` required: - id - status - format - platforms - created_at - completed_at - error - price_cents - charged - result additionalProperties: false required: - data additionalProperties: false '400': description: Validation error content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '401': description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '403': description: Forbidden content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '404': description: Not found content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '409': description: Conflict (incl. idempotency conflicts) content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '429': description: Rate limited content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' components: schemas: ErrorEnvelope: type: object properties: error: type: object required: - code - message properties: code: type: string message: type: string fields: type: object additionalProperties: type: string doc_url: type: string format: uri securitySchemes: bearerAuth: type: http scheme: bearer ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.