Documentation Index
Fetch the complete documentation index at: https://docs.sailresearch.com/llms.txt
Use this file to discover all available pages before exploring further.
Sail supports the OpenAI Responses API (/v1/responses). If you’re already using the OpenAI SDK, switching to Sail is mostly a base URL and API key change.
1. Get your API key
Sign up at the Sail dashboard and create an API key.
2. Make a request
Install the OpenAI SDK and point it at Sail. Create a response with background=True so the request returns a response_id immediately, then poll until completion:
import time
from openai import OpenAI
client = OpenAI(
base_url="https://api.sailresearch.com/v1",
api_key="YOUR_SAIL_API_KEY",
)
response = client.responses.create(
model="moonshotai/Kimi-K2.5",
input="Explain the key ideas behind transformers.",
max_output_tokens=1000,
background=True, # return immediately with ID to poll
)
print(f"Started {response.id}, waiting ...")
while response.status in {"in_progress", "queued"}:
time.sleep(1)
response = client.responses.retrieve(response.id)
if response.status != "completed":
raise RuntimeError(f"Error: {response.status}")
print(response.output_text)
cURL Example
URL=https://api.sailresearch.com/v1/responses
AUTH="Authorization: Bearer YOUR_SAIL_API_KEY"
# Create the response (returns immediately with an id)
RESPONSE_ID=$(curl -s $URL \
-H "$AUTH" \
-H "Content-Type: application/json" \
-d '{
"model": "moonshotai/Kimi-K2.5",
"input": "Explain the key ideas behind transformers.",
"background": true,
"metadata": { "completion_window": "standard" }
}' | jq -r '.id')
# Poll until completed
until [ "$(curl -s $URL/$RESPONSE_ID \
-H "$AUTH" | jq -r '.status')" = "completed" ]; do
sleep 1
done
curl -s $URL/$RESPONSE_ID -H "$AUTH" | jq
3. Next steps
