Skip to main content
POST
/
api
/
llm
/
train
Start a new LLM/VLM training job
curl --request POST \
  --url https://training-suite.uat-infer.shakticloud.ai/api/llm/train/ \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form 'training_config_json={
  "model_type": "vlm",
  "base_model": "AIDC-AI/Ovis1.6-Llama3.2-3B",
  "adapter": {
    "lora_r": 16
  },
  "quantization": {
    "bits": 4,
    "llm_int8_threshold": 6,
    "llm_int8_has_fp16_weight": false,
    "bnb_4bit_compute_dtype": "float16",
    "bnb_4bit_use_double_quant": true,
    "bnb_4bit_quant_type": "nf4"
  },
  "trainer": {
    "type": "finetune",
    "learning_rate": 0.0001,
    "batch_size": 1,
    "epochs": 1,
    "gradient_accumulation_steps": 16,
    "logging_steps": 1
  },
  "image_column": "url",
  "output_column": "caption",
  "prompt": {
    "template": "Could you please interpret the image and write a detailed caption?"
  }
}
' \
  --form experiment_name=my_first_vlm_job \
  --form csv_file='@example-file' \
  --form org=0bf00b43-430a-4ca3-a8b3-b13cc8dc6d4d
{
  "request_id": "<string>",
  "status": "<string>",
  "message": "<string>"
}

Documentation Index

Fetch the complete documentation index at: https://docs.shaktistudio.ai/llms.txt

Use this file to discover all available pages before exploring further.

Authorizations

​
Authorization
string
header
required

JWT token for authentication

Headers

​
Authorization
string
required

Bearer token for authentication and authorization.

Body

multipart/form-data
​
training_config_json
string
required

JSON-formatted string containing the model training configuration.

Example:

"{\n \"model_type\": \"vlm\",\n \"base_model\": \"AIDC-AI/Ovis1.6-Llama3.2-3B\",\n \"adapter\": {\n \"lora_r\": 16\n },\n \"quantization\": {\n \"bits\": 4,\n \"llm_int8_threshold\": 6,\n \"llm_int8_has_fp16_weight\": false,\n \"bnb_4bit_compute_dtype\": \"float16\",\n \"bnb_4bit_use_double_quant\": true,\n \"bnb_4bit_quant_type\": \"nf4\"\n },\n \"trainer\": {\n \"type\": \"finetune\",\n \"learning_rate\": 0.0001,\n \"batch_size\": 1,\n \"epochs\": 1,\n \"gradient_accumulation_steps\": 16,\n \"logging_steps\": 1\n },\n \"image_column\": \"url\",\n \"output_column\": \"caption\",\n \"prompt\": {\n \"template\": \"Could you please interpret the image and write a detailed caption?\"\n }\n}\n"

​
experiment_name
string
required

Name assigned to the training experiment.

Example:

"my_first_vlm_job"

​
csv_file
file
required

CSV or JSONL file containing the training dataset.

​
org
string
required

Organization ID associated with the training job.

Example:

"0bf00b43-430a-4ca3-a8b3-b13cc8dc6d4d"

Response

200 - application/json

Training job submitted successfully.

​
request_id
string

Unique identifier for the training job request

​
status
string

Initial status of the training job

​
message
string

Additional information about the job submission