Follow Discord
Sweep 22 Oct 2026 · 21:04Z Build v2.1.309 528 read Stable v2.1.308 Latest v2.1.308 Next v2.1.309 Feeds RSS JSON llms.txt llms-full.txt Unofficial
One change · api

Create a Dream changedapi/beta/dreams/create

Nearest release: v2.1.245, published an hour after this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.

Recorded here
Lines+49added
Lines−46removed
From line 1 where the diff opens
First seen 14 Aug 2026 this site's first read of the page
Recorded edits21to this page, all time

# Create a Dream ## Headers ## Body parameters ## Returns ## Example ### Response (200) ## Create a Dream ### Header Parameters ### Body Parameters ### Returns ### Example #### Response

The whole hunk

from line 1, old and new numbered
/
lines
from line 1
1---
2title: Create a Dream
3url: https://platform.claude.com/docs/en/api/beta/dreams/create
4---
1# Create a Dream
52 
6## Create a Dream
3**POST** `/v1/dreams`
74 
8**post** `/v1/dreams`
9 
105Create a Dream
116 
12### Header Parameters
7## Headers
138 
149- `"anthropic-beta": optional array of AnthropicBeta`
1510 
from line 82
8782 
8883 - `"mid-conversation-tool-changes-2026-07-01"`
8984 
90### Body Parameters
85## Body parameters
9186 
9287- `inputs: array of BetaDreamInput`
9388 
94 - `BetaDreamMemoryStoreInput object { memory_store_id, type }`
89 - `BetaDreamMemoryStoreInput object`
9590 
9691 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.
9792 
9893 - `memory_store_id: string`
9994 
95 minLength: 1
96 
10097 - `type: "memory_store"`
10198 
102 - `"memory_store"`
99 - `BetaDreamSessionsInput object`
103100 
104 - `BetaDreamSessionsInput object { session_ids, type }`
105 
106101 Input session transcripts the dream reads.
107102 
108103 - `session_ids: array of string`
from line 104
109104 
110105 - `type: "sessions"`
111106 
112 - `"sessions"`
113 
114107- `model: string or BetaDreamModelConfigParam`
115108 
116109 Model identifier and configuration applied to every pipeline stage.
from line 110
117110 
118111 - `string`
119112 
120 - `BetaDreamModelConfigParam object { id, speed }`
113 - `BetaDreamModelConfigParam object`
121114 
122115 Model identifier and configuration applied to every pipeline stage.
123116 
from line 118
125118 
126119 Model identifier, e.g. "claude-opus-5". 1-256 characters.
127120 
121 minLength: 1, maxLength: 256
122 
128123 - `speed: optional "standard" or "fast" or null`
129124 
130125 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.
from line 130
135130 
136131- `instructions: optional string or null`
137132 
133 minLength: 1, maxLength: 4096
134 
138135- `output_behavior: optional BetaOutputBehavior`
139136 
140137 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.
141138 
142 - `BetaOutputBehaviorCreateNew object { type }`
139 - `BetaOutputBehaviorCreateNew object`
143140 
144141 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.
145142 
146143 - `type: "create_new"`
147144 
148 - `"create_new"`
145 - `BetaOutputBehaviorUpdateExisting object`
149146 
150 - `BetaOutputBehaviorUpdateExisting object { memory_store_id, type }`
151 
152147 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.
153148 
154149 - `memory_store_id: string`
155150 
151 minLength: 1
152 
156153 - `type: "update_existing"`
157154 
158 - `"update_existing"`
155## Returns
159156 
160### Returns
157- `BetaDream object`
161158 
162- `BetaDream object { id, archived_at, created_at, 11 more }`
163 
164159 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.
165160 
166161 - `id: string`
from line 164
169164 
170165 A timestamp in RFC 3339 format
171166 
167 format: date-time
168 
172169 - `created_at: string`
173170 
174171 A timestamp in RFC 3339 format
175172 
173 format: date-time
174 
176175 - `ended_at: string or null`
177176 
178177 A timestamp in RFC 3339 format
179178 
179 format: date-time
180 
180181 - `error: BetaDreamError or null`
181182 
182183 Failure detail for a Dream whose `status` is `failed`.
from line 188
187188 
188189 - `inputs: array of BetaDreamInput`
189190 
190 - `BetaDreamMemoryStoreInput object { memory_store_id, type }`
191 - `BetaDreamMemoryStoreInput object`
191192 
192193 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.
193194 
194195 - `memory_store_id: string`
195196 
197 minLength: 1
198 
196199 - `type: "memory_store"`
197200 
198 - `"memory_store"`
201 - `BetaDreamSessionsInput object`
199202 
200 - `BetaDreamSessionsInput object { session_ids, type }`
201 
202203 Input session transcripts the dream reads.
203204 
204205 - `session_ids: array of string`
from line 206
205206 
206207 - `type: "sessions"`
207208 
208 - `"sessions"`
209 
210209 - `instructions: string or null`
211210 
212211 - `model: BetaDreamModelConfig`
from line 216
217216 
218217 Model identifier, e.g. "claude-opus-5". 1-256 characters.
219218 
219 minLength: 1, maxLength: 256
220 
220221 - `speed: optional "standard" or "fast"`
221222 
222223 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.
from line 230
229230 
230231 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.
231232 
232 - `BetaOutputBehaviorCreateNew object { type }`
233 - `BetaOutputBehaviorCreateNew object`
233234 
234235 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.
235236 
236237 - `type: "create_new"`
237238 
238 - `"create_new"`
239 - `BetaOutputBehaviorUpdateExisting object`
239240 
240 - `BetaOutputBehaviorUpdateExisting object { memory_store_id, type }`
241 
242241 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.
243242 
244243 - `memory_store_id: string`
245244 
245 minLength: 1
246 
246247 - `type: "update_existing"`
247248 
248 - `"update_existing"`
249 
250249 - `outputs: array of BetaDreamOutput`
251250 
252251 - `memory_store_id: string`
from line 252
253252 
254253 - `type: "memory_store"`
255254 
256 - `"memory_store"`
257 
258255 - `session_id: string or null`
259256 
260257 - `status: BetaDreamStatus`
from line 270
273270 
274271 - `type: "dream"`
275272 
276 - `"dream"`
277 
278273 - `usage: BetaDreamUsage`
279274 
280275 Cumulative token usage for the dream across every pipeline stage.
from line 278
283278 
284279 Total tokens used to create prompt-cache entries (sum of all TTL tiers).
285280 
281 format: int32
282 
286283 - `cache_read_input_tokens: number`
287284 
288285 Total tokens read from prompt cache.
289286 
287 format: int32
288 
290289 - `input_tokens: number`
291290 
292291 Total uncached input tokens consumed across every pipeline stage.
293292 
293 format: int32
294 
294295 - `output_tokens: number`
295296 
296297 Total output tokens generated across every pipeline stage.
297298 
298### Example
299 format: int32
299300 
300```http
301## Example
302 
303```bash
301304curl https://api.anthropic.com/v1/dreams \
302305 -H 'Content-Type: application/json' \
303306 -H 'anthropic-version: 2023-06-01' \
from line 317
314317 }'
315318```
316319 
317#### Response
320### Response (200)
318321 
319322```json
320323{
Feedback