Models
List All Available Models
GET/api/v1/user/model
List all available models in your account.
Get Model Details
GET/api/v1/user/model/{model_id}
Get details about a specific model.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| model_id | string | Yes | The unique identifier of the model |
Create a New Model
POST/api/v1/user/model
Create a new model (optionally with a dataset attached).
Delete a Model
DELETE/api/v1/user/model/{id}
Delete a model by ID.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| id | string | Yes | The unique identifier of the model to delete |
Conversations
List All Conversations
GET/api/v1/user/conversations
Retrieves all conversations for a specific model.
Query Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| model_id | string | Yes | The ID of the model to list conversations for |
Create a New Conversation
POST/api/v1/user/conversations
Creates a new conversation thread.
Rename a Conversation
PATCH/api/v1/user/conversations/{conversation_id}
Updates the title of an existing conversation.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| conversation_id | string | Yes | The ID of the conversation to rename |
Delete a Conversation
DELETE/api/v1/user/conversations/{conversation_id}
Permanently deletes a conversation and all its messages.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| conversation_id | string | Yes | The ID of the conversation to delete |
Get All Messages from a Conversation
GET/api/v1/user/conversations/{conversation_id}/messages
Retrieves all messages from a specific conversation thread.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| conversation_id | string | Yes | The ID of the conversation |
Datasets
List All Datasets
GET/api/v1/user/dataset
Retrieves a list of all available datasets.
Get Dataset Details
GET/api/v1/user/dataset/{id}
Retrieves detailed information about a specific dataset.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| id | string | Yes | The unique identifier of the dataset |
Create a New Dataset
POST/api/v1/user/dataset
Creates a new dataset by uploading files or providing URLs to scrape.
Request Body
| Parameter | Type | Required | Description |
|---|---|---|---|
| name | string | Yes | Name of the dataset |
| description | string | No | Description of the dataset |
| files | array | No | Array of files to upload |
| urls | array | No | Array of URLs to scrape for data |
Upload Files to Dataset
POST/api/v1/user/dataset/upload
Adds files to an existing dataset.
Request Body
Multipart form data with the following fields:| Parameter | Type | Required | Description |
|---|---|---|---|
| dataset_id | string | Yes | The ID of the dataset |
| file | file | Yes | The file to upload |
Delete a Dataset
DELETE/api/v1/user/dataset/{id}
Deletes a dataset. This can be a soft delete (by default), or a permanent delete.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| id | string | Yes | The unique identifier of the dataset |
Query Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| permanent | boolean | No | Set to true for permanent deletion. Default is false (soft delete) |
Evaluations
List All Evaluations
GET/api/v1/evaluations
List all evaluations with pagination and filtering.
Query Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| page | integer | No | Page number for pagination |
| per_page | integer | No | Number of results per page (default: 20) |
Create a New Evaluation
POST/api/v1/evaluations
Create a new evaluation for a model.
Request Body
| Parameter | Type | Required | Description |
|---|---|---|---|
| model_id | string | Yes | The ID of the model to evaluate |
| name | string | Yes | Name of the evaluation |
| config | object | No | Model-specific configuration for evaluation |
Get Evaluation Details
GET/api/v1/evaluations/{evaluation_id}
Get details about a specific evaluation.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| evaluation_id | string | Yes | The unique identifier of the evaluation |
Start Running an Evaluation
POST/api/v1/evaluations/{evaluation_id}/start
Start running an evaluation.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| evaluation_id | string | Yes | The unique identifier of the evaluation |
Run RAGAS Evaluation
POST/api/v1/evaluations/{evaluation_id}/run-ragas
Run RAGAS evaluation with benchmark data.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| evaluation_id | string | Yes | The unique identifier of the evaluation |
Update Evaluation Metadata
PUT/api/v1/evaluations/{evaluation_id}
Update an evaluation’s metadata.
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| evaluation_id | string | Yes | The unique identifier of the evaluation |
Request Body
| Parameter | Type | Required | Description |
|---|---|---|---|
| name | string | No | Updated name for the evaluation |
| description | string | No | Updated description |
Delete an Evaluation
DELETE/api/v1/evaluations/{evaluation_id}
Delete an evaluation (cannot delete running evaluations).
Path Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| evaluation_id | string | Yes | The unique identifier of the evaluation |
Check Evaluation Name Availability
GET/api/v1/evaluations/check-name
Check if an evaluation name is available.
Query Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| name | string | Yes | The evaluation name to check |
Inference
Perform direct inference queries without streaming, perfect for simple question-answering and batch processing workflows using knowledge graphs.Run a Single Query
POST/api/v1/user/inference
Run a single query with your knowledge graph.
Required Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| query | string | Yes | The question or query to process |
| model_id | integer | Yes | The ID of the model to use for inference |
Optional Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| conversation_id | string | No | Continue an existing conversation by providing its ID |
| thread_id | string | No | Thread identifier for managing conversation history |
| enable_web_search | boolean | No | Enable web search to enhance responses (default: false) |
| max_tokens | integer | No | Maximum response length (default: model-specific) |
| temperature | float | No | Response randomness from 0.0 to 1.0 (default: model-specific) |
| image_generation | boolean | No | Enable image generation in responses (default: false) |
| image_parameters | object | No | Image generation config: width, height, steps, guidance |
| persona | object | No | User persona configuration for customized responses |
| base_model | boolean | No | Bypass RAG and use base model directly, useful for evaluations (default: false) |
| session_id | string | No | Session tracking identifier |
| user_id | string | No | User identifier for tracking and personalization |
| tenant_id | string | No | Multi-tenancy identifier for organization-level isolation |
Advanced sampling overrides
These parameters provide fine-grained control over response generation and are typically used for specialized use cases.| Parameter | Type | Required | Description |
|---|---|---|---|
| top_k | integer | No | Top-K sampling: limits token selection to top K options |
| top_p | float | No | Nucleus sampling: cumulative probability threshold |
| response_length_preference | string | No | Response verbosity: “concise”, “balanced”, or “detailed” |
| creativity_level | float | No | Balance between creative and deterministic outputs |
| factual_precision | float | No | Preference for factual accuracy in responses |

