Ollama API

Load a model

POST

http://localhost:11434/api/chat

If the messages array is empty, the model will be loaded into memory.

Request Example

Shell

JavaScript

Java

Swift

curl --location --request POST 'http://localhost:11434/api/chat' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "llama3.2",
    "messages": [
        {
            "role": "user",
            "content": "Hello"
        }
    ]
}'

Response Example

{
  "model": "llama3.2",
  "created_at":"2024-09-12T21:17:29.110811Z",
  "message": {
    "role": "assistant",
    "content": ""
  },
  "done_reason": "load",
  "done": true
}

Request

Body Params application/json

model

enum<string>

optional

Allowed value:

llama3.2

Example:

llama3.2

messages

array [object]

optional

Examples

Responses

🟢200Success

application/json

Body

model

string

required

created_at

string

required

message

object

required

role

string

required

content

string

required

done_reason

string

required

done

boolean

required

Chat request (with tools)

Unload a model