Ollama

Chat with Ollama AI models

post

Chat with Ollama AI models.

Available Models: gpt-oss:20b, gpt-oss:120b, llama3.2:3b, deepseek-r1:8b

Required Parameters

Parameter	Type	Description
model	string	Model ID (see available models above)
messages	array	Array of message objects with role and content

Optional Parameters

Parameter	Type	Default	Description
think	string/bool	-	Reasoning mode: low/medium/high (gpt-oss), true/false (deepseek), not supported (llama3.2). Omit to use model default.
mode	string	auto	Routing: auto, opengpu (blockchain), direct (low-latency)
stream	boolean	false	Enable streaming (not yet supported)
options	object	-	Generation options (see below)

Options Object (optional)

Parameter	Type	Description
temperature	float	Sampling temperature (0.0-2.0)
top_k	integer	Top-k sampling (>=1)
top_p	float	Nucleus sampling (0.0-1.0)
num_ctx	integer	Context window size (128-131072)
num_predict	integer	Max tokens to generate (-1 = infinite)
seed	integer	Random seed for reproducibility
stop	array	Stop sequences
repeat_penalty	float	Repetition penalty (>=0)

Async Mode

Set async: true to get a task_address immediately and poll for results.

Header parameters

x-api-keyany ofOptional

stringOptional

or

nullOptional

Body

application/json

Ollama-compatible chat completion request.

Supports all standard Ollama parameters with explicit validation.

modelstringRequired

Model identifier. Available: gpt-oss:20b, gpt-oss:120b, llama3.2:3b, deepseek-r1:8b

Example: gpt-oss:120b

streamany ofOptional

Enable streaming responses (not yet supported)

Default: false

booleanOptional

or

nullOptional

optionsany ofOptional

Model generation options (temperature, top_k, etc.)

or

nullOptional

thinkany ofOptional

Reasoning mode. gpt-oss: 'low'/'medium'/'high', deepseek-r1: true/false, llama3.2: not supported. Omit to use model default.

booleanOptional

or

string · enumOptionalPossible values:

or

nullOptional

modeany ofOptional

Routing mode: 'auto' (intelligent), 'opengpu' (blockchain), 'direct' (low-latency)

Default: auto

string · enumOptionalPossible values:

or

nullOptional

asyncany ofOptional

Async mode: returns task_address immediately, poll /v2/tasks/{task_address} for result. Default: false (sync mode).

Default: false

booleanOptional

or

nullOptional

Responses

200

Successful Response

application/json

202

Task accepted (async mode). Poll the poll_url for status.

application/json

422

Validation Error

application/json

post

/v2/ollama/api/chat

HTTPcURLJavaScriptPython

POST /v2/ollama/api/chat HTTP/1.1
Content-Type: application/json
Accept: */*
Content-Length: 85

{
  "model": "gpt-oss:120b",
  "messages": [
    {
      "role": "user",
      "content": "Hello, how are you?"
    }
  ]
}

Minimal (Recommended)With Reasoning (think)With Generation OptionsDeepSeek with ReasoningLlama 3.2 (No Think)Force Blockchain (OpenGPU)All Options

{
  "message": {
    "role": "text",
    "content": "text",
    "ANY_ADDITIONAL_PROPERTY": "anything"
  },
  "done": true,
  "task_address": "text",
  "mode": "text",
  "ANY_ADDITIONAL_PROPERTY": "anything"
}

List available AI models

get

List all available AI models.

Returns all active Ollama models. Access control is enforced at request time via tier model_restrictions - this endpoint shows what's available.

Responses

200

Successful Response

application/json

get

/v2/ollama/api/models

HTTPcURLJavaScriptPython

GET /v2/ollama/api/models HTTP/1.1
Accept: */*

200

Successful Response

{
  "models": [
    {
      "name": "gpt-oss:20b",
      "modified_at": "text",
      "size": 0,
      "digest": "",
      "details": {
        "ANY_ADDITIONAL_PROPERTY": "anything"
      },
      "ANY_ADDITIONAL_PROPERTY": "anything"
    }
  ],
  "ANY_ADDITIONAL_PROPERTY": "anything"
}