Skip to main content

List Models

Retrieve a paginated list of available models and adapters, with filtering by object type, trainability, inference engine, service type, name, and architecture.

Endpoint​

GET https://api.platform.knowledgator.com/models/

Public Endpoint: This catalog endpoint is publicly accessible and does not require authentication. Do not pass KnowledgatorAPI-Key or Authorization: Bearer headers (sending auth headers causes 401 Unauthorized: auth.invalid_token).

Public Catalog vs. Your Custom Models
  • GET /models/ queries the public open models catalog (base models like comprehend-it or extract-it, and official adapters). Use these model IDs as the base model when starting fine-tuning.
  • Your Custom Trained Models: When you fine-tune a model, your custom models and runs are managed under /trainings/. To view your trained models and runs, use List Trainings (GET /trainings/).

Headers​

NameTypeRequiredDescription
Content-TypeStringNoapplication/json

Request Parameters​

ParameterTypeRequiredDescription
offsetIntegerNoPagination offset (default: 0, min: 0)
limitIntegerNoMaximum results per page (default: 100, max: 10000)
object_typeStringNoFilter by object type: adapter (LoRA adapter) or model (open model)
trainableStringNoFilter trainable models: adapter or model
inferenceStringNoFilter by supported inference engine (e.g. comprehend-it, extract-it)
nameStringNoFilter by model name (1-64 chars)
typeStringNoFilter by service type: ner, text-classification, relation-extraction
architectureStringNoFilter by underlying architecture (e.g. gliner, gliclass)
tagStringNoFilter by tag keyword
taskStringNoFilter by specific task
modalityStringNoFilter by modality (e.g. text)
include_baseBooleanNoInclude base models not trained by customer (default: true)

Response​

{
"models": [
{
"id": "comprehend-it",
"name": "comprehend-it",
"type": "text-classification",
"description": "Zero-shot text classification and reasoning model",
"num_of_parameters": "494M",
"training_model": true,
"training_adapter": true,
"inference": "comprehend-it",
"created_at": "2024-01-15T10:30:00Z",
"updated_at": "2024-01-15T10:30:00Z",
"is_base": true
}
],
"total": 1,
"offset": 0,
"limit": 100
}

Response Fields​

  • models: Array of model objects
    • id: Unique model identifier
    • name: Model name
    • type: Service type (ner, text-classification, etc.)
    • num_of_parameters: Number of parameters (e.g. "494M", "1.1B", or numeric)
    • training_model: Boolean indicating if full model training is supported
    • training_adapter: Boolean indicating if LoRA adapter training is supported
    • inference: Inference service compatibility
    • fine_tuned: Boolean indicating if this is a fine-tuned model/adapter
    • parent_model: Parent base model ID if fine-tuned
    • description: Full markdown model description
    • short_description: Brief summary description
    • created_at: ISO 8601 timestamp of creation
    • updated_at: ISO 8601 timestamp of last update
    • is_base: Whether this is a base model (not trained by customer)
  • total: Total number of models matching the filter
  • offset: Current pagination offset
  • limit: Current pagination limit

Error Codes​

Status CodeDescription
200OK – Request successful.
400Bad Request – Invalid input or validation failed.
401Unauthorized – Invalid API key or email verification required.
500Internal Server Error – Server-side issue.
503Service Unavailable – Service temporarily unavailable.

Example Usage​

SDK Usage​

💡 Dedicated SDK Guides: See Python SDK | TypeScript SDK

TypeScript/JavaScript​

import { createClient } from '@knowledgator/typescript-sdk';

const kg = createClient('YOUR_API_KEY');

// List all models
const result = await kg.finetuning.listModels();
console.log(result.models);

// List with filtering
const filtered = await kg.finetuning.listModels({
type: 'ner',
include_base: true,
limit: 50
});
console.log(filtered.models);

Python​

from knowledgator import Knowledgator

kg = Knowledgator(api_key="YOUR_API_KEY")

# List all models
result = kg.finetuning.list_models()
print(result.models)

# List with filtering
filtered = kg.finetuning.list_models(
type="ner",
include_base=True,
limit=50
)
print(filtered.models)

Expected SDK Output​

[
{
"id": "comprehend-it",
"name": "comprehend-it",
"type": "text-classification",
"description": "Zero-shot text classification and reasoning model",
"created_at": "2024-01-15T10:30:00Z",
"updated_at": "2024-01-15T10:30:00Z",
"is_base": true
}
]

Python​

import requests

url = "https://api.platform.knowledgator.com/models/"

# List all models
response = requests.get(url)
print(response.json())

# List with filtering (e.g. LoRA adapters trainable with comprehend-it)
params = {
"object_type": "adapter",
"trainable": "adapter",
"inference": "comprehend-it",
"limit": 50
}
response = requests.get(url, params=params)
print(response.json())

cURL​

# List all models (Public - no API key needed)
curl -X GET "https://api.platform.knowledgator.com/models/"

# List with filtering
curl -X GET "https://api.platform.knowledgator.com/models/?object_type=adapter&trainable=adapter&inference=comprehend-it&limit=50"

Expected Response (cURL)​

{
"models": [
{
"id": "comprehend-it",
"name": "comprehend-it",
"type": "text-classification",
"description": "Zero-shot text classification and reasoning model",
"created_at": "2024-01-15T10:30:00Z",
"updated_at": "2024-01-15T10:30:00Z",
"is_base": true
}
],
"total": 1,
"offset": 0,
"limit": 50
}

Additional Information​

  • Pagination: Use offset and limit parameters to paginate through large result sets
  • Filtering: Combine multiple filter parameters to narrow down results (e.g. by object_type, trainable, inference)
  • Base Models: Browse the public base models and adapters available as starting points for fine-tuning
  • Custom Trained Models: To manage models you have fine-tuned yourself, see List Trainings
  • Use Cases:
    • Browse available base models for fine-tuning
    • Find models by type, name, architecture, or task
    • Discover compatible inference models