Claude Platform Docs

Create a Dream

$ ant beta:dreams create
POST/v1/dreams

Create a Dream

Parameters
--input: array of BetaDreamInput

Body param

--model: string or BetaDreamModelConfigParam { id, speed }

Body param: Model identifier and configuration applied to every pipeline stage.

--instructions: optional string

Body param

minLength1
maxLength4096
--output-behavior: optional BetaOutputBehaviorCreateNew { type } or BetaOutputBehaviorUpdateExisting { memory_store_id, type }

Body param: The default destination: the job creates a new output memory store as a clone of the memory_store input and writes the consolidated memories into it. The input store is never mutated.

--beta: optional array of AnthropicBeta

Header param: Optional header to specify the beta version(s) you want to use.

Returns
beta_dream: object{ id, archived_at, created_at, 11 more }

An asynchronous memory-consolidation job that reads a memory store plus a set of session transcripts and writes consolidated memories into an output memory store — a new store by default, or an existing store chosen via output_behavior. The Dreams API is in research preview: the request and response shapes are volatile and may change without the deprecation period that applies to generally-available endpoints.

id: string
archived_at: string

A timestamp in RFC 3339 format

formatdate-time
created_at: string

A timestamp in RFC 3339 format

formatdate-time
ended_at: string

A timestamp in RFC 3339 format

formatdate-time
error: object{ message, type }

Failure detail for a Dream whose status is failed.

message: string
type: string
inputs: array of BetaDreamInput
One of the following:
beta_dream_memory_store_input: object{ memory_store_id, type }

An input memory store the dream reads from. The dream never mutates this store unless it is also the destination: with output_behavior {type: "update_existing"} the job consolidates this store in place.

memory_store_id: string
minLength1
type: "memory_store"
beta_dream_sessions_input: object{ session_ids, type }

Input session transcripts the dream reads.

session_ids: array of string
type: "sessions"
instructions: string
model: object{ id, speed }

Model identifier and configuration applied to every pipeline stage. Same wire shape as the Agents API ModelConfig.

id: string

Model identifier, e.g. "claude-opus-5". 1-256 characters.

minLength1
maxLength256
speed: optional "standard" or "fast"

Inference speed mode. fast provides significantly faster output token generation at premium pricing. Not all models support fast; invalid combinations are rejected at create time.

One of the following:
"standard"
"fast"
output_behavior: BetaOutputBehaviorCreateNew { type } or BetaOutputBehaviorUpdateExisting { memory_store_id, type }

The default destination: the job creates a new output memory store as a clone of the memory_store input and writes the consolidated memories into it. The input store is never mutated.

One of the following:
beta_output_behavior_create_new: object{ type }

The default destination: the job creates a new output memory store as a clone of the memory_store input and writes the consolidated memories into it. The input store is never mutated.

type: "create_new"
beta_output_behavior_update_existing: object{ memory_store_id, type }

The job writes the consolidated memories into this existing memory store instead of creating one. In EAP the store must be the job's own memory_store input, so the job consolidates the store in place.

memory_store_id: string
minLength1
type: "update_existing"
outputs: array of BetaDreamOutput { memory_store_id, type }
memory_store_id: string
type: "memory_store"
session_id: string
status: "pending" or "running" or "completed" or 2 more

Lifecycle status of a Dream.

One of the following:
"pending"
"running"
"completed"
"failed"
"canceled"
type: "dream"
usage: object{ cache_creation_input_tokens, cache_read_input_tokens, input_tokens, output_tokens }

Cumulative token usage for the dream across every pipeline stage.

cache_creation_input_tokens: number

Total tokens used to create prompt-cache entries (sum of all TTL tiers).

formatint32
cache_read_input_tokens: number

Total tokens read from prompt cache.

formatint32
input_tokens: number

Total uncached input tokens consumed across every pipeline stage.

formatint32
output_tokens: number

Total output tokens generated across every pipeline stage.

formatint32

Create a Dream

ant beta:dreams create \
  --api-key my-anthropic-api-key \
  --input '{memory_store_id: x, type: memory_store}' \
  --model string
{
  "id": "id",
  "archived_at": "2019-12-27T18:11:19.117Z",
  "created_at": "2019-12-27T18:11:19.117Z",
  "ended_at": "2019-12-27T18:11:19.117Z",
  "error": {
    "message": "message",
    "type": "type"
  },
  "inputs": [
    {
      "memory_store_id": "x",
      "type": "memory_store"
    }
  ],
  "instructions": "instructions",
  "model": {
    "id": "x",
    "speed": "standard"
  },
  "output_behavior": {
    "type": "create_new"
  },
  "outputs": [
    {
      "memory_store_id": "memory_store_id",
      "type": "memory_store"
    }
  ],
  "session_id": "session_id",
  "status": "pending",
  "type": "dream",
  "usage": {
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "input_tokens": 0,
    "output_tokens": 0
  }
}
Returns Examples
{
  "id": "id",
  "archived_at": "2019-12-27T18:11:19.117Z",
  "created_at": "2019-12-27T18:11:19.117Z",
  "ended_at": "2019-12-27T18:11:19.117Z",
  "error": {
    "message": "message",
    "type": "type"
  },
  "inputs": [
    {
      "memory_store_id": "x",
      "type": "memory_store"
    }
  ],
  "instructions": "instructions",
  "model": {
    "id": "x",
    "speed": "standard"
  },
  "output_behavior": {
    "type": "create_new"
  },
  "outputs": [
    {
      "memory_store_id": "memory_store_id",
      "type": "memory_store"
    }
  ],
  "session_id": "session_id",
  "status": "pending",
  "type": "dream",
  "usage": {
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "input_tokens": 0,
    "output_tokens": 0
  }
}