Pick the right action for every user request

Superlinked gives your agent one API to tag user requests with the actions your app supports, with a score for each action.

Get started

User request

Play the top twenty Hisham Abbas on Youtube

play music 0.981
add to playlist 0.016
search creative work 0.003
search screening event 0.000
get weather 0.000
rate book 0.000
book restaurant 0.000

Written for play music knowledgator/gliclass-large-v3.0

All 14 held-out requests land on the action they were written for

View evidence

add Highway Patrolman in my playlist this is al green
SNIPS AddToPlaylist validation row 61
1. add to playlist 1.000
2. search creative work 0.000
Written for add to playlist Ranked 1st
book a table at a Fried chicken restaurant
SNIPS BookRestaurant validation row 49
1. book restaurant 1.000
2. add to playlist 0.000
Written for book restaurant Ranked 1st
what is the overcast forecast for the current position starting on jul. 19, 2030
SNIPS GetWeather validation row 52
1. get weather 1.000
2. add to playlist 0.000
Written for get weather Ranked 1st
Play Subconscious Lobotomy from Jennifer Paull.
SNIPS PlayMusic validation row 9
1. play music 0.997
2. search creative work 0.003
Written for play music Ranked 1st
Give Small Screen, Big Picture a 0 out of 6 rating.
SNIPS RateBook validation row 24
1. rate book 1.000
2. search creative work 0.000
Written for rate book Ranked 1st
Looking for the picture with the name of Who Made Stevie Crye?
SNIPS SearchCreativeWork validation row 77
1. search creative work 1.000
2. add to playlist 0.000
Written for search creative work Ranked 1st
What is the nearest movie house with Window Connection playing at lunch
SNIPS SearchScreeningEvent validation row 27
1. search screening event 1.000
2. add to playlist 0.000
Written for search screening event Ranked 1st

Distinct label names decide the result. The 7 action names above never overlap, and all 14 recorded requests ranked their own action first; one request per action is drawn here and 7 more are counted. Where labels did overlap, on CFPB complaint products and CLINC150 assistant domains, the same model led on only 3 to 5 of 12 texts across 4 earlier runs. Read the four earlier runs

Score any text against your own labels in one call

View on GitHub

from sie_sdk import SIEClient
client = SIEClient(
    api_key="sk-sie-…",
    base_url="https://api.superlinked.com",
)
result = client.extract(
    "modelknowledgator/gliclass-large-v3.0",
    {"text": "textPlay the top twenty Hisham Abbas on Youtube"},
    labels=[
        "labeladd to playlist",
        "labelbook restaurant",
        "labelget weather",
        "labelplay music",
        "labelrate book",
        "labelsearch creative work",
        "labelsearch screening event",
    ],
)
print(result["classifications"])  # candidate label → confidence

import { SIEClient } from '@superlinked/sie-sdk';

const client = new SIEClient('https://api.superlinked.com', {
  apiKey: 'sk-sie-…',
});

const result = await client.extract(
  'knowledgator/gliclass-large-v3.0',
  { text: "Play the top twenty Hisham Abbas on Youtube" },
  { labels: ["add to playlist","book restaurant","get weather","play music","rate book","search creative work","search screening event"] },
);
console.log(result.classifications);

curl https://api.superlinked.com/v1/extract/knowledgator%2Fgliclass-large-v3.0 \
  -H "Authorization: Bearer sk-sie-…" \
  -H "Content-Type: application/json" \
  -d "{\"items\":[{\"text\":\"Play the top twenty Hisham Abbas on Youtube\"}],\"params\":{\"labels\":[\"add to playlist\",\"book restaurant\",\"get weather\",\"play music\",\"rate book\",\"search creative work\",\"search screening event\"]}}"

Build the "Classify" capability into my app using the Superlinked Inference Engine (SIE).

Context
- SIE is an OpenAI-style inference API. Python SDK: `from sie_sdk import SIEClient`; TypeScript: `@superlinked/sie-sdk`.
- Base URL: https://api.superlinked.com (or my regional endpoint). Auth: Bearer key from env `SIE_API_KEY` (never hard-code it).
- Model: knowledgator/gliclass-large-v3.0 (SIE primitive: /extract). Keep the model id configurable.

Task
- Input: a block of text plus a set of candidate labels.
- Behaviour: return the best-matching label(s) with confidence scores
- Call the selected SIE primitive once per request and map the response into your domain type.

Deliverables
- A typed client wrapper, an application-level function for this task, error handling for timeouts/empty input, and unit tests with a stubbed client.
- Wire it into my existing stack (ask me which framework if unclear) and add a short usage example.

Run

Output

play music0.981

add to playlist0.016

search creative work0.003

search screening event0.000

get weather0.000

rate book0.000

book restaurant0.000

Top label play music at 0.981 across 7 candidates

Zero-shot, no training data
You supply the candidate labels

PRICE

$ / 1M input tokens ↓

$0.124 SIE GLiClass Large

$0.2 OpenAI GPT-5.4 nano

$0.3 Google Gemini 3.5 Flash-Lite

OpenAI GPT-5.4 mini $0.75

Anthropic Claude Haiku 4.5 $1

QUALITY

Accuracy · AG News ↑

SIE GLiClass Large 0.74

0.72

0.71

0.70

LATENCY

p50 ms ↓

SIE GLiClass Large 70ms

82ms

88ms

90ms

95ms

Deploy your way

Managed Cloud

Full compute toolkit for your agents with zero ops.

No idle GPUs, pay for what you use
Fits your stack: SDK, API, CLI, MCP
Zero lock-in, self-host the same stack
SOC 2 Type 2, US or EU data residency

$50 Example credit balance

1.3B tokens embedded

120k pages parsed

27M tokens generated (1:1 in:out)

Get started

no credit card required

Self-host with K8s

Easy & scalable deployment in your own cloud.

Terraform to your cloud in minutes
Apache-2.0, same engine as Cloud
Scales to zero, no bill between jobs
Per-tenant pools, no noisy neighbors

Agent prompt

Deploy SIE to our AWS account with the superlinked/sie/aws Terraform module. Docs: superlinked.com/docs/deploymentDeploy SIE to our GCP project with the superlinked/sie/google Terraform module. Docs: superlinked.com/docs/deploymentDeploy SIE to our Azure AKS cluster via helm install. Requirements: superlinked.com/docs/deployment

Deploy guide

Run locally

Run the same models on your own machine.

Runs on NVIDIA GPU or Apple Silicon
One command, no Docker or cluster
All 100+ Cloud models, fully offline
Same SDK and IDs, no code changes

pip install "sie-server[local]" && sie-server servepip install "sie-server[local]" && sie-server serve --device cuda

Quickstart

All 14 held-out requests land on the action they were written for

Score any text against your own labels in one call

Classification quality and latency

Deploy your way