Documentation
Documentation
https://docs.inceptionlabs.ai
APIReference
https://docs.inceptionlabs.ai/api-reference
GettingStarted
https://docs.inceptionlabs.ai/get-started/get-started
OpenAPI Specification
openapi: 3.1.0
info:
title: Inception Chat Edit API
version: 1.0.0
description: Inception Labs LLM API — chat, FIM, and edit completions powered by Mercury diffusion language models. OpenAI-compatible request/response shapes with extensions for diffusion-specific features.
contact:
name: Inception Labs
url: https://docs.inceptionlabs.ai
email: support@inceptionlabs.ai
license:
name: Proprietary
x-logo:
url: https://docs.inceptionlabs.ai/logo.png
servers:
- url: https://api.inceptionlabs.ai
description: Production
security:
- BearerAuth: []
tags:
- name: Edit
description: Code edit completion endpoints.
paths:
/v1/edit/completions:
post:
summary: Create a code edit completion
description: Generate an edit to a code region given the surrounding context. The request must contain a single user message whose content includes the required edit prompt tags (e.g. `<|code_to_edit|>`, `<|cursor|>`, `<|current_file_content|>`). Returns an `EditCompletion` object. Streaming and tool calling are not supported on this endpoint.
operationId: createEditCompletion
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/EditCompletionRequest'
example:
model: mercury-edit-2
messages:
- role: user
content: "<|recently_viewed_code_snippets|>\n<|/recently_viewed_code_snippets|>\n<|edit_diff_history|>\n<|/edit_diff_history|>\n<|current_file_content|>\ndef greet(name):\n print(\"hi\")\n<|/current_file_content|>\n<|code_to_edit|>\ndef greet(name):\n print(\"hi\")<|cursor|>\n<|/code_to_edit|>"
required: true
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/EditCompletionResponse'
example:
id: cmpl-7a2b3c4d5e
object: edit.completion
created: 1745798400
model: mercury-edit-2
choices:
- index: 0
finish_reason: stop
message:
role: assistant
content: "def greet(name):\n print(f\"hi {name}\")\n"
usage:
prompt_tokens: 156
completion_tokens: 18
total_tokens: 174
reasoning_tokens: 0
cached_input_tokens: 0
'400':
description: Bad Request — invalid parameters.
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
example:
error:
message: You exceeded the maximum context length for this model of 128000. Please reduce the length of the messages or completion.
type: invalid_request_error
param: messages
code: context_length_exceeded
'401':
description: Unauthorized — missing or invalid API key.
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
example:
error:
message: Incorrect API key provided
type: authentication_error
param: null
code: invalid_api_key
'402':
description: Payment Required — billing inactive or quota exceeded.
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
example:
error:
message: Account is inactive
type: account_error
param: null
code: account_error
'404':
description: Not Found — model not available for this account.
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
example:
error:
message: model `jupyter-2` not found
type: invalid_request_error
param: model
code: model_not_found
'429':
description: Too Many Requests — rate limit exceeded.
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
example:
error:
message: Rate limit exceeded. Please try again later.
type: rate_limit_error
param: null
code: rate_limit_reached
'500':
description: Internal Server Error.
content:
application/json:
schema:
$ref: '#/components/schemas/ErrorResponse'
example:
error:
message: The server had an error while processing your request.
type: server_error
param: null
code: server_error
security:
- BearerAuth: []
tags:
- Edit
x-codeSamples:
- lang: python
label: Python
source: "import os\nfrom inceptionai import Inception\n\nclient = Inception(\n api_key=os.environ.get(\"INCEPTION_API_KEY\"), # defaults to this env var; can be omitted\n)\n\nedit_completion = client.edit.completions.create(\n model=\"mercury-edit-2\",\n messages=[\n {\n \"role\": \"user\",\n \"content\": \"<|recently_viewed_code_snippets|>\\n<|/recently_viewed_code_snippets|>\\n<|edit_diff_history|>\\n<|/edit_diff_history|>\\n<|current_file_content|>\\ndef greet(name):\\n print(\\\"hi\\\")\\n<|/current_file_content|>\\n<|code_to_edit|>\\ndef greet(name):\\n print(\\\"hi\\\")<|cursor|>\\n<|/code_to_edit|>\"\n }\n ],\n)\nprint(edit_completion)"
- lang: typescript
label: TypeScript
source: "import Inception from 'inceptionai';\n\nconst client = new Inception({\n apiKey: process.env['INCEPTION_API_KEY'], // defaults to this env var; can be omitted\n});\n\nconst editCompletion = await client.edit.completions.create({\n model: 'mercury-edit-2',\n messages: [\n {\n role: 'user',\n content: '<|recently_viewed_code_snippets|>\\n<|/recently_viewed_code_snippets|>\\n<|edit_diff_history|>\\n<|/edit_diff_history|>\\n<|current_file_content|>\\ndef greet(name):\\n print(\"hi\")\\n<|/current_file_content|>\\n<|code_to_edit|>\\ndef greet(name):\\n print(\"hi\")<|cursor|>\\n<|/code_to_edit|>'\n }\n ]\n});\nconsole.log(editCompletion);"
components:
schemas:
EditUsage:
properties:
prompt_tokens:
type: integer
title: Prompt Tokens
reasoning_tokens:
type: integer
title: Reasoning Tokens
completion_tokens:
type: integer
title: Completion Tokens
total_tokens:
type: integer
title: Total Tokens
cached_input_tokens:
type: integer
title: Cached Input Tokens
type: object
required:
- prompt_tokens
- reasoning_tokens
- completion_tokens
- total_tokens
- cached_input_tokens
title: EditUsage
description: Usage for edit completions.
EditMessage:
properties:
role:
type: string
const: assistant
title: Role
description: Edit responses are always from the assistant.
default: assistant
content:
anyOf:
- type: string
- items:
additionalProperties: true
type: object
type: array
- type: 'null'
title: Content
description: The model's edit output.
type: object
title: EditMessage
description: Edit response message. The model always responds as the assistant.
required:
- role
EditMessageParam:
properties:
role:
type: string
const: user
title: Role
description: Edit requests only accept user messages.
default: user
content:
anyOf:
- type: string
- items:
additionalProperties: true
type: object
type: array
title: Content
description: The content of the edit request. Must contain the required edit prompt tags.
type: object
required:
- content
- role
title: EditMessageParam
description: Edit request message. Edit endpoints only accept a single user message.
ErrorResponse:
type: object
title: ErrorResponse
required:
- error
properties:
error:
$ref: '#/components/schemas/ErrorObject'
example:
error:
message: You exceeded the maximum context length for this model of 128000. Please reduce the length of the messages or completion.
type: invalid_request_error
param: messages
code: context_length_exceeded
EditChoice:
properties:
index:
type: integer
title: Index
message:
$ref: '#/components/schemas/EditMessage'
finish_reason:
type:
- string
- 'null'
enum:
- stop
- length
- content_filter
- null
description: The reason the model stopped generating tokens.
type: object
required:
- index
- message
- finish_reason
title: EditChoice
description: Choice for edit completions.
EditCompletionResponse:
properties:
id:
type: string
title: Id
object:
type: string
const: edit.completion
title: Object
default: edit.completion
created:
type: integer
title: Created
model:
type: string
title: Model
choices:
items:
$ref: '#/components/schemas/EditChoice'
type: array
title: Choices
usage:
$ref: '#/components/schemas/EditUsage'
warning:
anyOf:
- type: string
- type: 'null'
title: Warning
type: object
required:
- id
- created
- model
- choices
- usage
- object
title: EditCompletionResponse
example:
id: cmpl-7a2b3c4d5e
object: edit.completion
created: 1745798400
model: mercury-edit-2
choices:
- index: 0
finish_reason: stop
message:
role: assistant
content: "def greet(name):\n print(f\"hi {name}\")\n"
usage:
prompt_tokens: 156
completion_tokens: 18
total_tokens: 174
reasoning_tokens: 0
cached_input_tokens: 0
EditCompletionRequest:
properties:
messages:
items:
$ref: '#/components/schemas/EditMessageParam'
type: array
title: Messages
model:
type: string
title: Model
description: The model to use for the edit completion.
max_tokens:
type: integer
title: Max Tokens
description: Maximum number of tokens to generate.
default: 1024
minimum: 1
maximum: 8192
temperature:
type: number
title: Temperature
description: What sampling temperature to use, between 0 and 2. Higher values make the output more random; lower values make it more focused and deterministic.
default: 0.2
minimum: 0.0
maximum: 2.0
top_p:
type: number
title: Top P
description: Float that controls the cumulative probability of the top tokens to consider.
default: 1.0
minimum: 0.0
maximum: 1.0
presence_penalty:
type: number
title: Presence Penalty
description: Number between -2 and 2. Positive values penalize tokens based on whether they have appeared in the text so far, increasing the model's likelihood to talk about new topics.
default: 1.0
minimum: -2.0
maximum: 2.0
type: object
required:
- messages
- model
title: EditCompletionRequest
ErrorObject:
type: object
title: ErrorObject
required:
- message
- type
properties:
message:
type: string
description: Human-readable error message.
type:
type: string
description: Error category, e.g. `invalid_request_error`, `rate_limit_error`.
param:
type:
- string
- 'null'
description: Offending parameter, if applicable.
code:
type:
- string
- 'null'
description: Machine-readable error code.
securitySchemes:
BearerAuth:
type: http
scheme: bearer
description: 'API key provided as a Bearer token: `Authorization: Bearer <api_key>`. Get an API key at https://platform.inceptionlabs.ai.'