Skip to main content
POST
https://api.classer.ai
/
v1
/
score
Score
curl --request POST \
  --url https://api.classer.ai/v1/score \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "source": "<string>",
  "attribute": "<string>",
  "description": "<string>",
  "model": "<string>"
}
'
{
  "score": 0.9234,
  "latency_ms": 125,
  "usage": {
    "prompt_tokens": 35,
    "completion_tokens": 1,
    "total_tokens": 36
  }
}

​
Request

​
source
string
required
The text to score
​
attribute
string
required
The attribute to score (e.g., “urgency”, “toxicity”, “quality”)
​
description
string
Optional description of the attribute for better accuracy
​
model
string
Model override (optional)

​
Response

​
score
number
Score between 0 and 1
​
latency_ms
number
Processing time in milliseconds
​
usage
object
Token usage information

​
Common Attributes

AttributeDescription
urgencyHow urgent is the message
toxicityLevel of toxic/offensive content
sentimentPositive sentiment (1) vs negative (0)
formalityHow formal is the writing
qualityContent quality
spamLikelihood of being spam

​
Examples

​
Urgency scoring

import classer

result = classer.score(
    source="URGENT! Server is down! Need help NOW!",
    attribute="urgency"
)

print(result.score)  # 0.95

​
Toxicity detection

# Toxic content
result = classer.score(
    source="You're an idiot and your product is garbage",
    attribute="toxicity"
)
print(result.score)  # 0.89

# Non-toxic content
result = classer.score(
    source="Thank you for your help, I really appreciate it!",
    attribute="toxicity"
)
print(result.score)  # 0.08

​
Custom attribute with description

result = classer.score(
    source="I've been waiting 3 weeks for my order",
    attribute="frustration",
    description="Level of customer frustration or dissatisfaction"
)

print(result.score)  # 0.72

​
Prioritizing support tickets

Python
import classer

tickets = [
    "When will the new feature be released?",
    "URGENT: Production database is corrupted!",
    "How do I change my email address?",
]

# Score urgency for each ticket
for ticket in tickets:
    result = classer.score(source=ticket, attribute="urgency")
    print(f"{result.score:.2f} - {ticket[:40]}...")

# Output:
# 0.15 - When will the new feature be released?...
# 0.98 - URGENT: Production database is corrupted...
# 0.12 - How do I change my email address?...

{
  "score": 0.9234,
  "latency_ms": 125,
  "usage": {
    "prompt_tokens": 35,
    "completion_tokens": 1,
    "total_tokens": 36
  }
}