You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit f68fe91
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: .speakeasy/in.openapi.yaml
+245-2Lines changed: 245 additions & 2 deletions
Original file line number
Diff line number
Diff line change
@@ -10299,6 +10299,15 @@ components:
10299
10299
example: 'https://contoso.openai.azure.com'
10300
10300
minLength: 1
10301
10301
type: 'string'
10302
+
declared_parameters:
10303
+
description: 'Parameters the upstream supports beyond the tested ones. Replaces the earlier declaration; tested parameters stay. Order and repeated names are ignored, but the list holds at most as many entries as there are declarable parameters, repeats included. `tool_choice` requires `tools` to be tested or declared.'
description: 'Attest where this deployment processes data.'
10304
10313
enum:
@@ -10315,6 +10324,12 @@ components:
10315
10324
type:
10316
10325
- 'boolean'
10317
10326
- 'null'
10327
+
max_completion_tokens:
10328
+
description: 'Cap on completion tokens per request. Requests asking for more skip this endpoint, and the default output length stays within it. When omitted, the cap is copied from the public endpoint for the same model and provider, if there is one.'
10329
+
example: 131072
10330
+
maximum: 2147483647
10331
+
minimum: 1
10332
+
type: 'integer'
10318
10333
model_permaslug:
10319
10334
description: 'Permanent slug of the model this endpoint serves.'
10320
10335
example: 'openai/gpt-4o-2024-08-06'
@@ -16039,6 +16054,12 @@ components:
16039
16054
type:
16040
16055
- 'boolean'
16041
16056
- 'null'
16057
+
max_completion_tokens:
16058
+
description: 'Most completion tokens a request routed to this endpoint can produce, or `null` when the endpoint sets no cap beyond the model context length.'
16059
+
example: 131072
16060
+
type:
16061
+
- 'integer'
16062
+
- 'null'
16042
16063
required:
16043
16064
- 'declared_zdr'
16044
16065
- 'declared_region'
@@ -25410,9 +25431,13 @@ components:
25410
25431
PrivateEndpoint:
25411
25432
example:
25412
25433
base_url: 'https://contoso.openai.azure.com'
25434
+
declared_parameters:
25435
+
- 'temperature'
25436
+
- 'seed'
25413
25437
declared_region: 'us'
25414
25438
declared_zdr: true
25415
25439
id: '5b1c4c4e-7d0a-4a8e-9f3a-2d6c1b0e8a11'
25440
+
max_completion_tokens: 131072
25416
25441
model_name: 'OpenAI: GPT-4o'
25417
25442
model_permaslug: 'openai/gpt-4o-2024-08-06'
25418
25443
model_slug: 'openai/gpt-4o'
@@ -25424,6 +25449,10 @@ components:
25424
25449
provider_name: 'Azure'
25425
25450
provider_slug: 'azure'
25426
25451
status: 'active'
25452
+
tested_parameters:
25453
+
- 'tools'
25454
+
- 'tool_choice'
25455
+
- 'response_format'
25427
25456
upstream_model_id: 'gpt-4o-prod'
25428
25457
properties:
25429
25458
base_url:
@@ -25432,6 +25461,14 @@ components:
25432
25461
type:
25433
25462
- 'string'
25434
25463
- 'null'
25464
+
declared_parameters:
25465
+
description: 'Parameters you declared the upstream supports. Requests that use a parameter only route to this endpoint when it is tested or declared.'
description: 'Most completion tokens a request routed to this endpoint can produce, or `null` when the endpoint sets no cap beyond the model context length.'
25487
+
example: 131072
25488
+
type:
25489
+
- 'integer'
25490
+
- 'null'
25448
25491
model_name:
25449
25492
description: 'Display name of the model, or `null` when the model is no longer listed.'
description: 'Parameters a validation run proved the upstream honors (`tools`, `tool_choice`, `response_format`, `structured_outputs`, `stop`). Read-only; cleared when the upstream configuration changes.'
25523
+
example:
25524
+
- 'tools'
25525
+
- 'tool_choice'
25526
+
- 'response_format'
25527
+
items:
25528
+
type: 'string'
25529
+
type: 'array'
25478
25530
upstream_model_id:
25479
25531
description: 'Model or deployment identifier sent to the upstream provider.'
25480
25532
example: 'gpt-4o-prod'
@@ -25492,6 +25544,8 @@ components:
25492
25544
- 'declared_zdr'
25493
25545
- 'declared_region'
25494
25546
- 'pricing'
25547
+
- 'tested_parameters'
25548
+
- 'declared_parameters'
25495
25549
type: 'object'
25496
25550
PrivateEndpointActivation:
25497
25551
description: 'Validate and activate in the same call. On a failed validation the draft is kept and returned with a 422, so fix it and call `/validate` and `/activate` instead of creating it again.'
@@ -25559,6 +25613,36 @@ components:
25559
25613
- 'validation_stale'
25560
25614
example: 'no_byok_key'
25561
25615
type: 'string'
25616
+
PrivateEndpointDeclaredParameter:
25617
+
description: 'A chat request parameter the endpoint accepts.'
25618
+
enum:
25619
+
- 'temperature'
25620
+
- 'top_p'
25621
+
- 'top_k'
25622
+
- 'min_p'
25623
+
- 'top_a'
25624
+
- 'frequency_penalty'
25625
+
- 'presence_penalty'
25626
+
- 'repetition_penalty'
25627
+
- 'max_tokens'
25628
+
- 'max_completion_tokens'
25629
+
- 'logit_bias'
25630
+
- 'logprobs'
25631
+
- 'top_logprobs'
25632
+
- 'prediction'
25633
+
- 'seed'
25634
+
- 'response_format'
25635
+
- 'structured_outputs'
25636
+
- 'stop'
25637
+
- 'tools'
25638
+
- 'tool_choice'
25639
+
- 'parallel_tool_calls'
25640
+
- 'reasoning'
25641
+
- 'reasoning_effort'
25642
+
- 'web_search_options'
25643
+
- 'verbosity'
25644
+
example: 'seed'
25645
+
type: 'string'
25562
25646
PrivateEndpointDeclaredRegion:
25563
25647
description: 'Where you attest this deployment processes data. `global` means not regional; `null` means undeclared.'
25564
25648
enum:
@@ -25570,6 +25654,20 @@ components:
25570
25654
type:
25571
25655
- 'string'
25572
25656
- 'null'
25657
+
PrivateEndpointParameterCheck:
25658
+
example:
25659
+
parameter: 'response_format'
25660
+
supported: true
25661
+
properties:
25662
+
parameter:
25663
+
example: 'response_format'
25664
+
type: 'string'
25665
+
supported:
25666
+
type: 'boolean'
25667
+
required:
25668
+
- 'parameter'
25669
+
- 'supported'
25670
+
type: 'object'
25573
25671
PrivateEndpointPricing:
25574
25672
description: 'Negotiated per-token rates reported for requests routed to this endpoint.'
25575
25673
example:
@@ -25665,6 +25763,11 @@ components:
25665
25763
items:
25666
25764
$ref: '#/components/schemas/PrivateEndpointCheck'
25667
25765
type: 'array'
25766
+
parameters:
25767
+
description: 'Request parameters probed after the checks passed, each with whether the upstream honored it. Supported ones are recorded in `tested_parameters`. Absent on a failed run and for providers that run no parameter probes; empty when every probe group was already proven.'
description: 'Request parameters probed after the checks passed, each with whether the upstream honored it. Supported ones are recorded in `tested_parameters`. Absent on a failed run and for providers that run no parameter probes; empty when every probe group was already proven.'
description: 'Parameters the upstream supports beyond the tested ones. Replaces the earlier declaration; tested parameters stay. Order and repeated names are ignored, but the list holds at most as many entries as there are declarable parameters, repeats included. `tool_choice` requires `tools` to be tested or declared.'
description: 'Create a private endpoint as a draft. Drafts are not routable: validate one with `POST /private-endpoints/{id}/validate`, then activate it with `POST /private-endpoints/{id}/activate`. Pass `activate` to do all three in one call; if validation fails the draft is kept and returned with the failed checks. Send an `Idempotency-Key` header to make retries safe: a repeated key with the same request returns the endpoint the first request created; reusing it with different fields returns 422 `idempotency_key_reused`. [Management key](/docs/guides/overview/auth/management-api-keys) required.'
46097
46219
operationId: 'createPrivateEndpoint'
46098
46220
parameters:
46099
-
- description: 'Retry-safe create: a repeated create with the same key from the same organization returns the endpoint the first request created instead of creating another. The endpoint is returned as it is now. Reusing a key with a different request body (model, provider, base URL, upstream model ID, declared ZDR or region, or pricing) is rejected with 422 `idempotency_key_reused`. `activate` is not compared, and later edits to the endpoint do not affect the comparison.'
46221
+
- description: 'Retry-safe create: a repeated create with the same key from the same organization returns the endpoint the first request created instead of creating another. The endpoint is returned as it is now. Reusing a key with a different request body (model, provider, base URL, upstream model ID, declared ZDR or region, pricing, or declared parameters) is rejected with 422 `idempotency_key_reused`. Declared parameters are compared as a set, so a retry that reorders or repeats a name replays the endpoint. `activate` and `max_completion_tokens` are not compared, and later edits to the endpoint do not affect the comparison.'
46100
46222
in: 'header'
46101
46223
name: 'idempotency-key'
46102
46224
required: false
46103
46225
schema:
46104
-
description: 'Retry-safe create: a repeated create with the same key from the same organization returns the endpoint the first request created instead of creating another. The endpoint is returned as it is now. Reusing a key with a different request body (model, provider, base URL, upstream model ID, declared ZDR or region, or pricing) is rejected with 422 `idempotency_key_reused`. `activate` is not compared, and later edits to the endpoint do not affect the comparison.'
46226
+
description: 'Retry-safe create: a repeated create with the same key from the same organization returns the endpoint the first request created instead of creating another. The endpoint is returned as it is now. Reusing a key with a different request body (model, provider, base URL, upstream model ID, declared ZDR or region, pricing, or declared parameters) is rejected with 422 `idempotency_key_reused`. Declared parameters are compared as a set, so a retry that reorders or repeats a name replays the endpoint. `activate` and `max_completion_tokens` are not compared, and later edits to the endpoint do not affect the comparison.'
46105
46227
example: 'wayfair-gpt-4o-eastus-2026-09'
46106
46228
maxLength: 255
46107
46229
minLength: 1
@@ -46867,6 +46989,127 @@ paths:
46867
46989
tags:
46868
46990
- 'Private Endpoints'
46869
46991
x-speakeasy-name-override: 'enable'
46992
+
/private-endpoints/{id}/parameters:
46993
+
put:
46994
+
description: 'Declare which request parameters the upstream supports beyond the ones validation proved. Applies to drafts and active endpoints; routing picks the change up for subsequent requests. The body replaces the earlier declaration and is read as a set: order and repeated names do not matter, but the list holds at most as many entries as there are declarable parameters, repeats included. `tested_parameters` stay. Returns 400 `parameters_invalid` when `tool_choice` is declared without `tools` being tested or declared. [Management key](/docs/guides/overview/auth/management-api-keys) required.'
46995
+
operationId: 'updatePrivateEndpointParameters'
46996
+
parameters:
46997
+
- description: 'Stable identifier of the private endpoint.'
46998
+
in: 'path'
46999
+
name: 'id'
47000
+
required: true
47001
+
schema:
47002
+
description: 'Stable identifier of the private endpoint.'
description: 'Set the negotiated per-token rates reported for requests routed to this endpoint. Applies to drafts and active endpoints; new rates take effect for subsequent requests. The body replaces the whole pricing object: omitting `cache_read` or `cache_write` reverts that rate to the provider pricing strategy''s default, even if it was set before. [Management key](/docs/guides/overview/auth/management-api-keys) required.'
0 commit comments