Training Status

Get the current training status including indexes being trained and the first-train vector-count floor.

Authentication

Required - admin (root) API key via X-API-Key header. Per-user (cdbk_) keys are rejected with 403:

X-API-Key: cyborg_your_api_key_here

When RBAC is not enabled (CYBORGDB_SERVICE_ROOT_KEY unset), no header is required.

Request

This is a GET request with no request body. Authentication is required via the admin API key header.

Response

{
  "training_indexes": ["my_index", "semantic_index"],
  "retrain_threshold": 65536,
  "currently_training": "my_index",
  "queued_indexes": ["semantic_index"],
  "worker_running": true,
  "worker_pid": 0,
  "global_training": {}
}

Show response fields

array[string]

List of index names currently being trained or queued for training

integer

Vector-count floor before the first automatic training: an untrained index is auto-trained once num_vectors > retrain_threshold. Reflects the service’s AUTO_TRAIN_MIN_VECTORS setting (default 65536).

string | null

Name of the index currently being trained, or null if none

array[string]

List of indexes queued for training (waiting their turn)

boolean

Whether the training worker thread is active

integer

(Deprecated) Always 0. Kept for SDK backward compatibility

object

(Deprecated) Always empty object. Kept for SDK backward compatibility

Training States

Training state is inferred from the currently_training and queued_indexes fields:

If currently_training is non-null, that index is actively being trained
Indexes in queued_indexes are waiting their turn
An index not in either field has completed or has not been trained

Exceptions

401 Unauthorized

Authentication failed (invalid API key)

403 Forbidden

Authenticated with a per-user (cdbk_) key — training status requires the admin (root) API key

500 Internal Server Error

Internal server error

Example Usage

Get Training Status:

curl -X GET "http://localhost:8000/v1/indexes/training-status" \
     -H "X-API-Key: cyborg_your_api_key_here"

Monitor Training Progress:

# Check if any indexes are being trained
training_count=$(curl -s -X GET "http://localhost:8000/v1/indexes/training-status" \
     -H "X-API-Key: cyborg_your_api_key_here" | \
     jq '.training_indexes | length')

if [ "$training_count" -gt 0 ]; then
    echo "$training_count indexes are currently being trained"
fi

Wait for Training Completion:

# Poll until specific index training is complete
index_name="my_index"

while true; do
    is_training=$(curl -s -X GET "http://localhost:8000/v1/indexes/training-status" \
         -H "X-API-Key: cyborg_your_api_key_here" | \
         jq --arg idx "$index_name" '.training_indexes | contains([$idx])')
    
    if [ "$is_training" = "false" ]; then
        echo "Training complete for $index_name"
        break
    fi
    
    echo "Still training $index_name..."
    sleep 5
done

Check First-Train Floor:

# Get the current first-train vector-count floor
threshold=$(curl -s -X GET "http://localhost:8000/v1/indexes/training-status" \
     -H "X-API-Key: cyborg_your_api_key_here" | \
     jq '.retrain_threshold')

echo "Untrained indexes will auto-train once they exceed ${threshold} vectors"

Training is automatically triggered when:

An untrained index’s vector count exceeds the first-train floor (num_vectors > retrain_threshold)
A trained index grows enough that its optimal number of inverted lists exceeds the current n_lists
Manual training is requested via the /v1/indexes/train endpoint

Set AUTO_TRAIN_DISABLED=true on the service to turn off automatic training. See Train an Encrypted Index for details.

Use this endpoint to avoid querying an index while it’s being trained, as training can temporarily affect query performance.

Introduction

Client

Encrypted Index

Types & Getters

Users (RBAC)

Authentication

Request

Response

Training States

Exceptions

Example Usage

​Authentication

​Request

​Response

​Training States

​Exceptions

​Example Usage

Authentication

Request

Response

Training States

Exceptions

Example Usage