Ollama provides compatibility with parts of the OpenAI API, allowing existing applications built for OpenAI to connect to locally-running models through Ollama. Supported endpoints include chat completions, completions, embeddings, models, and the Responses API.
openapi: 3.1.0
info:
title: Ollama OpenAI Compatibility API
description: >-
Ollama provides compatibility with parts of the OpenAI API, allowing
existing applications built for OpenAI to connect to locally-running
models through Ollama. Supported endpoints include chat completions,
completions, embeddings, models, images, and the Responses API.
version: '0.1.0'
contact:
name: Ollama Team
url: https://ollama.com
license:
name: MIT
url: https://opensource.org/licenses/MIT
externalDocs:
description: Ollama OpenAI Compatibility Documentation
url: https://docs.ollama.com/api/openai-compatibility
servers:
- url: http://localhost:11434/v1
description: Local Ollama Server (OpenAI-compatible)
tags:
- name: Chat Completions
description: >-
Generate chat completions using the OpenAI-compatible chat endpoint
with multi-turn conversation support.
- name: Completions
description: >-
Generate text completions using the OpenAI-compatible completions
endpoint.
- name: Embeddings
description: >-
Generate vector embeddings using the OpenAI-compatible embeddings
endpoint.
- name: Images
description: >-
Generate images from text descriptions using the OpenAI-compatible
images endpoint. Experimental feature.
- name: Models
description: >-
List and retrieve model information using the OpenAI-compatible
models endpoints.
- name: Responses
description: >-
Generate structured responses with optional reasoning using the
OpenAI-compatible Responses API.
security:
- bearerAuth: []
paths:
/chat/completions:
post:
operationId: createChatCompletion
summary: Ollama Create chat completion
description: >-
Creates a model response for the given chat conversation. Supports
streaming, JSON mode, structured output, vision, and tool calling.
Compatible with the OpenAI Chat Completions API format.
tags:
- Chat Completions
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/ChatCompletionRequest'
responses:
'200':
description: Successful chat completion response
content:
application/json:
schema:
$ref: '#/components/schemas/ChatCompletionResponse'
text/event-stream:
schema:
$ref: '#/components/schemas/ChatCompletionStreamResponse'
'400':
description: Bad Request
'404':
description: Model not found
/completions:
post:
operationId: createCompletion
summary: Ollama Create completion
description: >-
Creates a completion for the provided prompt. Compatible with the
OpenAI Completions API format. Supports streaming, JSON mode,
and reproducible outputs.
tags:
- Completions
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/CompletionRequest'
responses:
'200':
description: Successful completion response
content:
application/json:
schema:
$ref: '#/components/schemas/CompletionResponse'
text/event-stream:
schema:
$ref: '#/components/schemas/CompletionStreamResponse'
'400':
description: Bad Request
'404':
description: Model not found
/embeddings:
post:
operationId: createEmbedding
summary: Ollama Create embeddings
description: >-
Creates an embedding vector representing the input text. Compatible
with the OpenAI Embeddings API format.
tags:
- Embeddings
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/EmbeddingRequest'
responses:
'200':
description: Successful embedding response
content:
application/json:
schema:
$ref: '#/components/schemas/EmbeddingResponse'
'400':
description: Bad Request
'404':
description: Model not found
/models:
get:
operationId: listModels
summary: Ollama List models
description: >-
Lists the currently available models and provides basic information
about each one. Compatible with the OpenAI List Models API format.
tags:
- Models
responses:
'200':
description: Successful response with list of models
content:
application/json:
schema:
$ref: '#/components/schemas/ModelListResponse'
/models/{model}:
get:
operationId: retrieveModel
summary: Ollama Retrieve a model
description: >-
Retrieves a model instance, providing basic information about the
model. Compatible with the OpenAI Retrieve Model API format.
tags:
- Models
parameters:
- name: model
in: path
required: true
description: >-
The identifier of the model to retrieve.
schema:
type: string
responses:
'200':
description: Successful response with model information
content:
application/json:
schema:
$ref: '#/components/schemas/ModelObject'
'404':
description: Model not found
/images/generations:
post:
operationId: createImage
summary: Ollama Create image
description: >-
Creates an image given a text prompt. This is an experimental
endpoint that may change or be removed. Only supports
b64_json response format.
tags:
- Images
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/ImageGenerationRequest'
responses:
'200':
description: Successful image generation response
content:
application/json:
schema:
$ref: '#/components/schemas/ImageGenerationResponse'
'400':
description: Bad Request
'404':
description: Model not found
/responses:
post:
operationId: createResponse
summary: Ollama Create response
description: >-
Creates a structured response with optional reasoning. Supports
streaming, tool calling, and reasoning summaries. Compatible
with the OpenAI Responses API format. Added in Ollama v0.13.3.
tags:
- Responses
requestBody:
required: true
content:
application/json:
schema:
$ref: '#/components/schemas/ResponseRequest'
responses:
'200':
description: Successful response
content:
application/json:
schema:
$ref: '#/components/schemas/ResponseObject'
text/event-stream:
schema:
$ref: '#/components/schemas/ResponseStreamEvent'
'400':
description: Bad Request
'404':
description: Model not found
components:
securitySchemes:
bearerAuth:
type: http
scheme: bearer
description: >-
API key authentication. The key is accepted but not validated
by Ollama. Use any value such as ollama.
schemas:
ChatCompletionRequest:
type: object
description: >-
Request body for creating a chat completion in OpenAI-compatible
format.
required:
- model
- messages
properties:
model:
type: string
description: >-
The model to use for chat completion.
messages:
type: array
description: >-
A list of messages comprising the conversation so far.
items:
$ref: '#/components/schemas/ChatCompletionMessage'
temperature:
type: number
description: >-
Sampling temperature between 0 and 2. Higher values make
output more random.
minimum: 0.0
maximum: 2.0
top_p:
type: number
description: >-
Nucleus sampling parameter. Considers tokens with top_p
probability mass.
minimum: 0.0
maximum: 1.0
max_tokens:
type: integer
description: >-
Maximum number of tokens to generate in the response.
frequency_penalty:
type: number
description: >-
Penalty for token frequency to reduce repetition.
minimum: -2.0
maximum: 2.0
presence_penalty:
type: number
description: >-
Penalty for token presence to encourage topic diversity.
minimum: -2.0
maximum: 2.0
seed:
type: integer
description: >-
Random seed for deterministic generation.
stop:
description: >-
Sequences where the API will stop generating further tokens.
oneOf:
- type: string
- type: array
items:
type: string
stream:
type: boolean
description: >-
If true, partial message deltas are sent as server-sent
events.
default: false
stream_options:
type: object
description: >-
Options for streaming responses.
properties:
include_usage:
type: boolean
description: >-
If true, includes usage information in the stream.
response_format:
type: object
description: >-
Specifies the format of the response. Use type json_object
for JSON mode or json_schema for structured output.
properties:
type:
type: string
description: >-
The response format type.
enum:
- text
- json_object
- json_schema
json_schema:
type: object
description: >-
The JSON Schema for structured output.
additionalProperties: true
tools:
type: array
description: >-
A list of tools the model may call.
items:
$ref: '#/components/schemas/OpenAIToolDefinition'
ChatCompletionMessage:
type: object
description: >-
A message in the chat conversation.
required:
- role
properties:
role:
type: string
description: >-
The role of the message author.
enum:
- system
- user
- assistant
- tool
content:
description: >-
The content of the message. Can be a string or an array
of content parts for multimodal input.
oneOf:
- type: string
- type: array
items:
$ref: '#/components/schemas/ContentPart'
name:
type: string
description: >-
An optional name for the participant.
tool_calls:
type: array
description: >-
Tool calls generated by the model.
items:
$ref: '#/components/schemas/OpenAIToolCall'
tool_call_id:
type: string
description: >-
Tool call that this message is responding to.
ContentPart:
type: object
description: >-
A content part within a multimodal message.
required:
- type
properties:
type:
type: string
description: >-
The type of content part.
enum:
- text
- image_url
text:
type: string
description: >-
The text content when type is text.
image_url:
type: object
description: >-
The image URL or base64 data when type is image_url.
properties:
url:
type: string
description: >-
URL of the image or a base64-encoded data URI.
ChatCompletionResponse:
type: object
description: >-
Response object from a chat completion request.
properties:
id:
type: string
description: >-
A unique identifier for the chat completion.
object:
type: string
description: >-
The object type, always chat.completion.
const: chat.completion
created:
type: integer
description: >-
Unix timestamp of when the completion was created.
model:
type: string
description: >-
The model used for the completion.
choices:
type: array
description: >-
A list of chat completion choices.
items:
$ref: '#/components/schemas/ChatCompletionChoice'
usage:
$ref: '#/components/schemas/UsageStats'
ChatCompletionChoice:
type: object
description: >-
A single chat completion choice.
properties:
index:
type: integer
description: >-
The index of the choice in the list.
message:
$ref: '#/components/schemas/ChatCompletionMessage'
finish_reason:
type: string
description: >-
The reason the model stopped generating tokens.
enum:
- stop
- length
- tool_calls
ChatCompletionStreamResponse:
type: object
description: >-
A streaming chat completion chunk.
properties:
id:
type: string
description: >-
A unique identifier for the chat completion.
object:
type: string
description: >-
The object type, always chat.completion.chunk.
const: chat.completion.chunk
created:
type: integer
description: >-
Unix timestamp of when the chunk was created.
model:
type: string
description: >-
The model used for the completion.
choices:
type: array
description: >-
A list of chat completion chunk choices.
items:
type: object
properties:
index:
type: integer
description: >-
The index of the choice.
delta:
type: object
description: >-
The delta content for this chunk.
properties:
role:
type: string
description: >-
The role of the author.
content:
type: string
description: >-
The content delta.
tool_calls:
type: array
description: >-
Tool call deltas.
items:
$ref: '#/components/schemas/OpenAIToolCall'
finish_reason:
type:
- string
- 'null'
description: >-
The finish reason, if applicable.
CompletionRequest:
type: object
description: >-
Request body for creating a text completion in OpenAI-compatible
format.
required:
- model
- prompt
properties:
model:
type: string
description: >-
The model to use for completion.
prompt:
description: >-
The prompt to generate completions for.
oneOf:
- type: string
- type: array
items:
type: string
temperature:
type: number
description: >-
Sampling temperature between 0 and 2.
minimum: 0.0
maximum: 2.0
top_p:
type: number
description: >-
Nucleus sampling parameter.
minimum: 0.0
maximum: 1.0
max_tokens:
type: integer
description: >-
Maximum number of tokens to generate.
frequency_penalty:
type: number
description: >-
Penalty for token frequency.
minimum: -2.0
maximum: 2.0
presence_penalty:
type: number
description: >-
Penalty for token presence.
minimum: -2.0
maximum: 2.0
seed:
type: integer
description: >-
Random seed for deterministic generation.
stop:
description: >-
Stop sequences for generation.
oneOf:
- type: string
- type: array
items:
type: string
stream:
type: boolean
description: >-
Whether to stream back partial progress.
default: false
stream_options:
type: object
description: >-
Options for streaming responses.
properties:
include_usage:
type: boolean
description: >-
If true, includes usage info in the stream.
suffix:
type: string
description: >-
The suffix that comes after the completion.
CompletionResponse:
type: object
description: >-
Response object from a text completion request.
properties:
id:
type: string
description: >-
A unique identifier for the completion.
object:
type: string
description: >-
The object type, always text_completion.
const: text_completion
created:
type: integer
description: >-
Unix timestamp of when the completion was created.
model:
type: string
description: >-
The model used for the completion.
choices:
type: array
description: >-
A list of completion choices.
items:
type: object
properties:
index:
type: integer
description: >-
The index of the choice.
text:
type: string
description: >-
The generated text.
finish_reason:
type: string
description: >-
The reason the model stopped generating.
enum:
- stop
- length
usage:
$ref: '#/components/schemas/UsageStats'
CompletionStreamResponse:
type: object
description: >-
A streaming text completion chunk.
properties:
id:
type: string
description: >-
A unique identifier for the completion.
object:
type: string
description: >-
The object type.
created:
type: integer
description: >-
Unix timestamp of creation.
model:
type: string
description: >-
The model used.
choices:
type: array
description: >-
A list of completion chunk choices.
items:
type: object
properties:
index:
type: integer
description: >-
The index of the choice.
text:
type: string
description: >-
The text delta.
finish_reason:
type:
- string
- 'null'
description: >-
The finish reason, if applicable.
EmbeddingRequest:
type: object
description: >-
Request body for creating embeddings in OpenAI-compatible format.
required:
- model
- input
properties:
model:
type: string
description: >-
The model to use for embedding generation.
input:
description: >-
Input text to embed. Can be a string or array of strings.
oneOf:
- type: string
- type: array
items:
type: string
encoding_format:
type: string
description: >-
The format of the returned embeddings.
enum:
- float
- base64
dimensions:
type: integer
description: >-
The number of dimensions for the output embeddings.
EmbeddingResponse:
type: object
description: >-
Response object from an embedding request.
properties:
object:
type: string
description: >-
The object type, always list.
const: list
data:
type: array
description: >-
The list of embedding objects.
items:
type: object
properties:
object:
type: string
description: >-
The object type, always embedding.
const: embedding
index:
type: integer
description: >-
The index of the embedding in the list.
embedding:
type: array
description: >-
The embedding vector.
items:
type: number
model:
type: string
description: >-
The model used to generate the embeddings.
usage:
$ref: '#/components/schemas/UsageStats'
ModelListResponse:
type: object
description: >-
Response containing a list of available models in OpenAI format.
properties:
object:
type: string
description: >-
The object type, always list.
const: list
data:
type: array
description: >-
The list of model objects.
items:
$ref: '#/components/schemas/ModelObject'
ModelObject:
type: object
description: >-
Information about a model in OpenAI-compatible format.
properties:
id:
type: string
description: >-
The model identifier.
object:
type: string
description: >-
The object type, always model.
const: model
created:
type: integer
description: >-
Unix timestamp when the model was last modified.
owned_by:
type: string
description: >-
The organization that owns the model. Defaults to library.
ImageGenerationRequest:
type: object
description: >-
Request body for generating images in OpenAI-compatible format.
This is an experimental endpoint.
required:
- model
- prompt
properties:
model:
type: string
description: >-
The model to use for image generation.
prompt:
type: string
description: >-
A text description of the desired image.
size:
type: string
description: >-
The size of the generated image, such as 1024x1024.
response_format:
type: string
description: >-
The format in which the image is returned. Only b64_json
is supported.
enum:
- b64_json
ImageGenerationResponse:
type: object
description: >-
Response object from an image generation request.
properties:
created:
type: integer
description: >-
Unix timestamp of when the image was created.
data:
type: array
description: >-
The generated image data.
items:
type: object
properties:
b64_json:
type: string
description: >-
The base64-encoded image data.
ResponseRequest:
type: object
description: >-
Request body for the Responses API in OpenAI-compatible format.
required:
- model
- input
properties:
model:
type: string
description: >-
The model to use for generating the response.
input:
description: >-
The input to generate a response for. Can be a string or
an array of input items.
oneOf:
- type: string
- type: array
items:
type: object
additionalProperties: true
instructions:
type: string
description: >-
System-level instructions for the model.
tools:
type: array
description: >-
A list of tools the model may call.
items:
$ref: '#/components/schemas/OpenAIToolDefinition'
stream:
type: boolean
description: >-
Whether to stream the response.
default: false
temperature:
type: number
description: >-
Sampling temperature between 0 and 2.
minimum: 0.0
maximum: 2.0
top_p:
type: number
description: >-
Nucleus sampling parameter.
minimum: 0.0
maximum: 1.0
max_output_tokens:
type: integer
description: >-
Maximum number of output tokens to generate.
ResponseObject:
type: object
description: >-
Response object from the Responses API.
properties:
id:
type: string
description: >-
A unique identifier for the response.
object:
type: string
description: >-
The object type.
created_at:
type: integer
description: >-
Unix timestamp of when the response was created.
model:
type: string
description: >-
The model used.
output:
type: array
description: >-
The output items generated by the model.
items:
type: object
additionalProperties: true
usage:
$ref: '#/components/schemas/UsageStats'
ResponseStreamEvent:
type: object
description: >-
A streaming event from the Responses API.
properties:
type:
type: string
description: >-
The type of streaming event.
data:
type: object
description: >-
The event data payload.
additionalProperties: true
OpenAIToolDefinition:
type: object
description: >-
A tool definition in OpenAI-compatible format.
required:
- type
- function
properties:
type:
type: string
description: >-
The type of tool. Currently only function is supported.
enum:
- function
function:
type: object
description: >-
The function definition.
required:
- name
properties:
name:
type: string
description: >-
The name of the function.
description:
type: string
description: >-
A description of what the function does.
parameters:
type: object
description: >-
The function parameters as a JSON Schema object.
additionalProperties: true
OpenAIToolCall:
type: object
description: >-
A tool call generated by the model.
properties:
id:
type: string
description: >-
A unique identifier for the tool call.
type:
type: string
description: >-
The type of tool call.
enum:
- function
function:
type: object
description: >-
The function call details.
properties:
name:
type: string
description: >-
The name of the function to call.
arguments:
type: string
description: >-
The arguments to pass to the function as a JSON string.
UsageStats:
type: object
description: >-
Token usage statistics for the request.
properties:
prompt_tokens:
type: integer
description: >-
Number of tokens in the prompt.
completion_tokens:
type: integer
description: >-
Number of tokens in the generated completion.
total_tokens:
type: integer
description: >-
Total number of tokens used in the request.