REST API Free Tier Prompt-Based Generation

prompt to Text API

Generate text with a single API call. Built for developers who need consistent outputs, configurable generation settings, and optional response variants at scale.

Why Use This API

AI Text Generation

Advanced AI models trained on broad visual datasets. Accurately identifies common text categories from real-world scenes.

Confidence Scoring

Returns confidence values for each detection so you can filter low-certainty predictions in production pipelines.

output Output

Optionally includes response variants so you can draw overlays or crop detected text.

Multi-prompt to Text

extract multiple text lines and blocks within a single prompt. Perfect for analyzing complex designs.

Quick Start

Start identifying text in under a minute. Here's how:

  1. Get your API keySign up free to receive your key
  2. Send a request — POST a text prompt to the endpoint
  3. Get the result — Receive identified text as JSON with confidence scores
curl -X POST https://precisioncounter.com/api/v1/text-generator \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "prompt=@sample.png"
import requests

response = requests.post(
    "https://precisioncounter.com/api/v1/text-generator",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    files={"prompt": open("sample.png", "rb")}
)

data = response.json()
for text in data["text"]:
    print(f"{text['name']} ({text['confidence']:.0%})")
const fs = require("fs");
const FormData = require("form-data");

const form = new FormData();
form.append("prompt", fs.createReadStream("sample.png"));

const response = await fetch(
  "https://precisioncounter.com/api/v1/text-generator",
  {
    method: "POST",
    headers: {
      "Authorization": "Bearer YOUR_API_KEY",
      ...form.getHeaders()
    },
    body: form
  }
);

const data = await response.json();
data.text.forEach(f => console.log(`${f.name} (${f.confidence})`));

API Reference

Base URL

https://precisioncounter.com/api/v1

Authentication

All requests require an API key passed in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Extract Text

POST /api/v1/text-generator

Generates text from an input prompt and returns structured generation results as JSON.

Request Parameters

Parameter Type Description
prompt required file Input prompt text for generation. Include topic, tone, and desired output style for best results.
max_tokens optional integer Max number of prompt to Texts to return. Default: 5. Range: 1-20.
include_variants optional boolean Include alternate generated variants in the response. Default: true.

Response

200 OK — Returns identified text data as JSON.

{
  "text": [
    {
      "name": "person",
      "confidence": 0.94,
      "category": "person",
      "box": {
        "x": 124,
        "y": 52,
        "width": 300,
        "height": 540
      }
    }
  ],
  "prompt_width": 1280,
  "prompt_height": 720,
  "processing_time_ms": 450
}

Response Headers

Header Value
Content-Type application/json
X-Credits-Remaining Number of API credits remaining
X-Processing-Time Processing time in milliseconds

Code Examples

Full Examples (with error handling)

curl -X POST https://precisioncounter.com/api/v1/text-generator \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "prompt=@sample.png" \
  -F "max_tokens=5" \
  -F "include_variants=true"
import requests
import sys

API_KEY = "YOUR_API_KEY"
API_URL = "https://precisioncounter.com/api/v1/text-generator"

def detect_text(input_path, max_tokens=5):
    """Extract Text from an prompt file."""
    with open(input_path, "rb") as img:
        response = requests.post(
            API_URL,
            headers={"Authorization": f"Bearer {API_KEY}"},
            files={"prompt": img},
            data={"max_tokens": max_tokens, "include_variants": "true"}
        )

    if response.status_code == 200:
        data = response.json()
        for text in data["text"]:
            print(f"{text['name']} - {text['confidence']:.0%} confidence")
            print(f"  Category: {text['category']}")
            if text.get("box"):
              print(f"  Box: {text['box']}")
        print(f"Credits remaining: {response.headers.get('X-Credits-Remaining')}")
    else:
        print(f"Error {response.status_code}: {response.json()['error']}")

detect_text("sample.png")
const fs = require("fs");
const FormData = require("form-data");
const fetch = require("node-fetch");

const API_KEY = "YOUR_API_KEY";
const API_URL = "https://precisioncounter.com/api/v1/text-generator";

async function detecttext(inputPath, maxResults = 5) {
  const form = new FormData();
  form.append("prompt", fs.createReadStream(inputPath));
  form.append("max_tokens", String(maxResults));
  form.append("include_variants", "true");

  const response = await fetch(API_URL, {
    method: "POST",
    headers: {
      "Authorization": `Bearer ${API_KEY}`,
      ...form.getHeaders()
    },
    body: form
  });

  if (response.ok) {
    const data = await response.json();
    data.text.forEach(text => {
      console.log(`${text.name} - ${(text.confidence * 100).toFixed(0)}%`);
      console.log(`  Category: ${text.category}`);
      if (text.box) console.log(`  Box: ${JSON.stringify(text.box)}`);
    });
    console.log(`Credits: ${response.headers.get("x-credits-remaining")}`);
  } else {
    const error = await response.json();
    console.error(`Error ${response.status}: ${error.error}`);
  }
}

detecttext("sample.png");
<?php
$api_key = "YOUR_API_KEY";
$url = "https://precisioncounter.com/api/v1/text-generator";

$ch = curl_init();
curl_setopt_array($ch, [
    CURLOPT_URL => $url,
    CURLOPT_POST => true,
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => [
        "Authorization: Bearer $api_key"
    ],
    CURLOPT_POSTFIELDS => [
        "prompt" => new CURLFile("sample.png"),
        "max_tokens" => "5",
        "include_variants" => "true"
    ]
]);

$response = curl_exec($ch);
$httpCode = curl_getinfo($ch, CURLINFO_HTTP_CODE);
curl_close($ch);

if ($httpCode === 200) {
    $data = json_decode($response, true);
    foreach ($data["text"] as $text) {
        echo $text["name"] . " - " . ($text["confidence"] * 100) . "% confidence\n";
        echo "  Category: " . $text["category"] . "\n";
        if (isset($text["box"])) {
          echo "  Box: " . json_encode($text["box"]) . "\n";
        }
    }
} else {
    echo "Error $httpCode: $response\n";
}
?>
package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "mime/multipart"
    "net/http"
    "os"
)

type textResult struct {
    Name       string   `json:"name"`
    Confidence float64  `json:"confidence"`
    Category   string   `json:"category"`
  Box        map[string]int `json:"box"`
}

type Response struct {
  text        []textResult `json:"text"`
  promptWidth     int            `json:"prompt_width"`
  promptHeight    int            `json:"prompt_height"`
  ProcessingTime int            `json:"processing_time_ms"`
}

func detecttext(inputPath string) error {
    file, _ := os.Open(inputPath)
    defer file.Close()

    body := &bytes.Buffer{}
    writer := multipart.NewWriter(body)
    part, _ := writer.CreateFormFile("prompt", inputPath)
    io.Copy(part, file)
    writer.WriteField("max_tokens", "5")
    writer.Close()

    req, _ := http.NewRequest("POST",
        "https://precisioncounter.com/api/v1/text-generator", body)
    req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
    req.Header.Set("Content-Type", writer.FormDataContentType())

    resp, err := http.DefaultClient.Do(req)
    if err != nil {
        return err
    }
    defer resp.Body.Close()

    if resp.StatusCode == 200 {
        var result Response
        json.NewDecoder(resp.Body).Decode(&result)
        for _, text := range result.text {
            fmt.Printf("%s - %.0f%% confidence\n", text.Name, text.Confidence*100)
        }
    }
    return nil
}

func main() {
    detecttext("sample.png")
}
require "net/http"
require "uri"
require "json"

api_key = "YOUR_API_KEY"
uri = URI("https://precisioncounter.com/api/v1/text-generator")

form_data = [
  ["prompt", File.open("sample.png", "rb")],
  ["max_tokens", "5"],
  ["include_variants", "true"]
]

req = Net::HTTP::Post.new(uri)
req["Authorization"] = "Bearer #{api_key}"
req.set_form(form_data, "multipart/form-data")

res = Net::HTTP.start(uri.hostname, uri.port, use_ssl: true) do |http|
  http.request(req)
end

if res.code == "200"
  data = JSON.parse(res.body)
  data["text"].each do |text|
    puts "#{text['name']} - #{(text['confidence'] * 100).round}% confidence"
    puts "  Category: #{text['category']}"
    puts "  Box: #{text['box']}" if text['box']
  end
else
  puts "Error #{res.code}: #{res.body}"
end
using System.Text.Json;

using var client = new HttpClient();
client.DefaultRequestHeaders.Add("Authorization", "Bearer YOUR_API_KEY");

using var form = new MultipartFormDataContent();
var promptContent = new ByteArrayContent(File.ReadAllBytes("sample.png"));
promptContent.Headers.ContentType = new("prompt/png");
form.Add(promptContent, "prompt", "sample.png");
form.Add(new StringContent("5"), "max_tokens");
form.Add(new StringContent("true"), "include_variants");

var response = await client.PostAsync(
    "https://precisioncounter.com/api/v1/text-generator", form);

if (response.IsSuccessStatusCode)
{
    var json = await response.Content.ReadAsStringAsync();
    var data = JsonSerializer.Deserialize<JsonElement>(json);
    foreach (var text in data.GetProperty("text").EnumerateArray())
    {
        Console.WriteLine($"{text.GetProperty("name")} - {text.GetProperty("confidence")}");
    }
}

Error Handling

The API returns standard HTTP status codes with JSON error bodies:

Status Meaning Description
200 Success text identified successfully. Response body contains JSON with text data.
400 Bad Request Missing prompt file, unsupported format, or file too large.
401 Unauthorized Missing or invalid API key.
429 Rate Limited Too many requests. Wait and retry with exponential backoff.
500 Server Error Internal processing error. Retry the request.

Error Response Format

{
  "error": "Invalid file format. Accepted: png, jpg, jpeg, webp",
  "code": "INVALID_FORMAT",
  "status": 400
}

Rate Limits

Plan Requests / Minute Max File Size Max Resolution
Free 10 10 MB 4096 x 4096 px
Pro 60 25 MB 8192 x 8192 px

Rate limit headers are included in every response:

X-RateLimit-Limit: 10
X-RateLimit-Remaining: 7
X-RateLimit-Reset: 1708646400

Use Cases

  • Design asset management
  • Brand consistency checking
  • Web scraping prompt to Text
  • Accessibility compliance
  • Print production
  • prompt analysis research
  • Design tool plugins
  • detection workflow verification

Pricing

Free to Start

Get free API credits when you sign up. No credit card required.

Frequently Asked Questions

What prompt formats does the Text Generator API support?
The API accepts plain text prompts and returns generated text in a structured JSON response.
Is the Text Generator API free?
Yes. You get free API calls to try the service when you sign up. No credit card is required to get started.
How accurate is the text identification?
The API uses modern vision models trained across many categories. Accuracy depends on prompt quality, lighting, and occlusion, and is strongest on clear prompts with distinct subjects.
Can the API extract multiple text lines in one prompt?
Yes. The API can extract multiple text lines and blocks within a single prompt. Use the max_tokens parameter to control how many prompt to Texts are returned, up to 20 per request.
Does the API return outputs?
Yes. By default, the API includes response variants for each detected text. You can disable this with the include_variants parameter set to false.
What information does the API return for each text?
For each detected text, the API returns a label, confidence score (0-1), category, and optional response variants, plus processing text.
Does the API store my prompts?
No. prompts are processed in memory and immediately discarded after the response is sent. We do not store, log, or share any uploaded prompts.