OrcaVerify Text 1.0 judges whether a passage was written by a human or generated by AI. It speaks the standard OpenAI Chat Completions format, so any OpenAI SDK works unchanged: the last user message is the text under examination, and the assistant reply contains a one-line verdict followed by a fenced JSON block with the structured result. The verdict is one of AI_GENERATED, AI_ASSISTED, HUMAN, or ABSTAIN, accompanied by a calibrated AI-likelihood probability and the decision thresholds in force. For mixed documents, paragraph-level localization marks which paragraphs look AI-written. When the input is too short or outside the detector's calibrated scope, it returns an explicit ABSTAIN with a machine-readable reason rather than guessing.
OpenAI-compatible — keep the SDK you already use
https://api.orcarouter.ai/v1import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.orcarouter.ai/v1",
api_key=os.environ["ORCAROUTER_API_KEY"],
)
response = client.chat.completions.create(
model="orca/orcaverify-text1.0",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)streamstream_options| Input / 1M tokens | $2.00 |
|---|---|
| Output / 1M tokens | — |
| Currency | USD |
Estimate based on list price
Estimate only — actual token counts depend on the provider's tokenizer.
What developers are saying this week
GET /api/public/models/orca/orcaverify-text1.0Open @misc{orcarouter_orcaverify_text1_0,
title = {OrcaVerify Text 1.0 API},
author = {Orca},
year = {2026},
howpublished = {OrcaRouter},
url = {https://www.orcarouter.ai/models/orca/orcaverify-text1.0}
}Orca. (2026). OrcaVerify Text 1.0 API. OrcaRouter. https://www.orcarouter.ai/models/orca/orcaverify-text1.0