---
title: "Get Predictions. Returns a 424 status code if the model is not a Sequence Classification model"
method: POST
path: "/predict"
tags: ["Text Embeddings Inference"]
---

# Get Predictions. Returns a 424 status code if the model is not a Sequence Classification model

`POST /predict`

## Request body

- PredictRequest
  - `inputs` union, required — Model input. Can be either a single string, a pair of strings or a batch of mixed single and pairs of strings.
    - string — A single string
    - string[] — A pair of strings
    - union[] — A batch
      - union
        - string[] — A single string
        - string[] — A pair of strings
  - `raw_scores` boolean
  - `truncate` boolean, nullable
  - `truncation_direction` 'Left' | 'Right'

## Response `200`

Predictions

- union
  - Prediction[]
    - `label` string, required
    - `score` number, float, required
  - array[]
    - Prediction[]
      - `label` string, required
      - `score` number, float, required

## Other responses

- `400` — Batch is empty
- `413` — Batch size error
- `422` — Tokenization error
- `424` — Prediction Error
- `429` — Model is overloaded

---

[API](https://skmtc.net/huggingface/apis/text-embeddings-inference.md) · [All operations](https://skmtc.net/huggingface/apis/text-embeddings-inference/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/huggingface/text-embeddings-inference/versions/f4c15add6d2b/schema)