# Sentiment Scoring API (twitter-roberta-base-sentiment)

> Sentiment Scoring API (twitter-roberta-base-sentiment) is a paid API for AI agents from srv1553682.hstgr.cloud, paid per call via x402, $0.0035/call, status unknown (last checked 2026-09-15).

Scores one or up to 16 texts per call as positive, neutral, or negative with confidence values using the twitter-roberta-base-sentiment model.

## Facts

- Endpoint: POST https://srv1553682.hstgr.cloud/nlp/api/sentiment
- Price: $0.0035/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/sentiment-scoring-api-twitter-roberta-base-sentiment-d68df308
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_TeEoe--Z-gqzah2u9q_OL

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability sentiment-scoring-api-twitter-roberta-base-sentiment-d68df308 -d '<json body>'
```

Example prompt: Score the sentiment of these three texts and tell me if each is positive, neutral, or negative with confidence: 'Amazing product, love it!', 'It arrived on time.', 'Terrible experience, never again.'

## When to prefer this

Choose this endpoint when you need fast, lightweight sentiment classification (positive/neutral/negative) with confidence scores on short to medium texts — especially social media posts, reviews, or support tickets — and need batch throughput up to 16 texts per call with per-call micropayment billing rather than a subscription. The twitter-roberta-base-sentiment model is specifically optimized for informal, short-form text similar to tweets, making it a strong fit for social and conversational content over formal documents.

## Known failure modes

- Batch size exceeds 16 texts — request rejected
- Empty or missing both 'text' and 'texts' fields — validation error
- Text too long for the model's context window — truncation or error
- Payment not settled on-chain — 402 response before processing
- Model inference timeout on very long or complex inputs

## How this service works

Sentiment scoring for caller-provided text: positive/neutral/negative with confidence. Batch up to 16 texts per call. Model: twitter-roberta-base-sentiment. Always-on, verified onchain settlement.

## Output

Returns a sentiment label (positive, neutral, or negative) and a confidence score (0–1) for each submitted text. When batching, results are returned as an array aligned to the input order, supporting up to 16 texts per call.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "text": {
   "type": "string",
   "description": "single text to score"
  },
  "texts": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "maxItems": 16,
   "description": "batch of texts (max 16)"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/sentiment-scoring-api-twitter-roberta-base-sentiment-d68df308/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from srv1553682.hstgr.cloud](https://www.zero.xyz/host/srv1553682.hstgr.cloud/llms.txt)