Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/No-Country-simulation/G9-LATAM-Team-58/llms.txt

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

The /model endpoint surfaces metadata about the machine learning model that is currently powering Mindloom’s classification and embedding pipeline. This is a thin proxy: the API fetches model information from the inference service’s internal /model/info endpoint and returns a trimmed summary. Use this endpoint to verify which model artifact is running, confirm the embedding dimension before performing vector operations, or display model quality metrics in a developer dashboard.

​
Endpoint

GET /model
Base URL: http://localhost:8080
Auth: None required
Query parameters: None

​
How it works

The Spring API controller delegates to IModelService, which calls the inference service at http://inference:8000/model/info and maps the full ModelInfoResponse payload to the smaller ModelResponse object returned here. If the inference service is unavailable, the request fails with a 503.

​
Response — 200 OK

A successful request returns a ModelResponse object.
​
version
string
required
Model artifact version string (e.g. "v1").
​
embeddingModel
string
required
HuggingFace model identifier used to generate embeddings. Currently "intfloat/multilingual-e5-small". This value originates from the model artifact metadata stored at training time.
​
dim
number
required
Embedding vector dimension. Currently 384, corresponding to the output size of intfloat/multilingual-e5-small.
​
macroF1
number
required
Macro-averaged F1 score measured during model evaluation, expressed as a value between 0 and 1. A higher score indicates better balanced classification performance across all categories.

​
Example

​
Request

curl http://localhost:8080/model

​
Response

{
  "version": "v1",
  "embeddingModel": "intfloat/multilingual-e5-small",
  "dim": 384,
  "macroF1": 0.87
}

​
Error Codes

HTTP Statuserror codeDescription
503INTERNAL_ERRORThe inference service is unreachable. Ensure the inference container is running and accessible at http://inference:8000.

​
503 error response

{
  "error": "INTERNAL_ERROR",
  "message": "Inference service unavailable"
}

Build docs developers (and LLMs) love

Get started for freeTalk to us