---
title: GLiNER2.5 Decide on SIE | Superlinked
description: SIE, an open-source inference engine, serves GLiNER2.5 Decide, an open classifier that scores every category you send in one call, with no prompt to write.
canonical_url: https://superlinked.com/models/fastino-gliner2-5-decide
last_updated: 2026-11-07
---

# GLiNER2.5 Decide

An open classifier that scores every category you send in one call, with no prompt to write.

Model id: `fastino/GLiNER2.5-Decide` · Endpoint: `extract` · [Hugging Face](https://huggingface.co/fastino/GLiNER2.5-Decide)

Listed under [Classify text](/models#classify) on [the model index](/models.md).

| Accuracy | Time per request | Serverless price |
|---|---|---|
| – (AG News, labelled correctly · not measured) | 13.1 ms per message | $0.28 per 1M tokens |

## On the task pages

- [Classify text 74x faster than GPT‑6 Luna, 85% cheaper](/classify): 87.5% Labelled correctly · 400 messages; 13.1 ms Time per message; $0.0035 Cost per 1,000 messages

## Call it

On SIE Serverless:

```python
from sie_sdk import SIEClient

client = SIEClient(api_key="sk-sie-…")

result = client.extract(
    "fastino/GLiNER2.5-Decide",
    {"text": "I've been charged more than once for the same transaction"},
    labels=[
        "pending card payment",
        "reverted card payment",
        "request a card payment refund",
        "duplicate card payment",
        "card payment fee",
        "unrecognized card payment",
        "declined card payment",
        "card not working",
    ],
)
print(result["classifications"])
```

## Next to the other zero-shot classifiers

| Model | AG News, labelled correctly | Time per request | Serverless price | Size |
|---|---|---|---|---|
| [GLiNER2.5 Decide](/models/fastino-gliner2-5-decide.md) `fastino/GLiNER2.5-Decide` | – | 13.1 ms per message | $0.28 per 1M tokens | 486M |
| [DeBERTa v3 large zero-shot v2.0](/models/moritzlaurer-deberta-v3-large-zeroshot-v2-0.md) `MoritzLaurer/deberta-v3-large-zeroshot-v2.0` | 88.6% | – | $0.129 per 1M tokens | 435M |
| [deberta v3 base zeroshot v2.0](/models/moritzlaurer-deberta-v3-base-zeroshot-v2-0.md) `MoritzLaurer/deberta-v3-base-zeroshot-v2.0` | 88.1% | – | Not on Serverless (run in your cloud) | 184M |
| [ModernBERT base zeroshot v2.0](/models/moritzlaurer-modernbert-base-zeroshot-v2-0.md) `MoritzLaurer/ModernBERT-base-zeroshot-v2.0` | 84.3% | – | Not on Serverless (run in your cloud) | 150M |
| [gliclass instruct large v1.0](/models/knowledgator-gliclass-instruct-large-v1-0.md) `knowledgator/gliclass-instruct-large-v1.0` | 76.9% | – | $0.19 per 1M tokens | 439M |

– = not measured / not timed / not listed.

[All 11 zero-shot classifiers](/models#classify)

## Facts

- Size: 486M parameters
- License: Apache-2.0
- Reads: Text, up to 512 tokens
- Endpoint: extract

## Evaluation

- Accuracy: Run on SIE against the public dataset named in each figure
- Time: SIE timed on a GPU server with the client alongside, one request at a time
- Prices: Serverless list prices
- Sources: [Task page runs and their inputs](https://huggingface.co/datasets/superlinked/sie-task-evidence)