Skip to main content
Sail serves trillions of tokens, with support for the best open-source models and your own LoRA fine-tunes. To achieve maximum efficiency for long-horizon agents, we serve traffic at higher latencies in tiers of service called completion windows. 
from openai import OpenAI

client = OpenAI(
    base_url="https://api.sailresearch.com/v1",
    api_key="YOUR_SAIL_API_KEY",
)

completion = client.chat.completions.create(
    model="zai-org/GLM-5.2-FP8",
    messages=[{"role": "user", "content": "What are the top 3 things to do in San Francisco?"}],
)

print(completion.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.sailresearch.com/v1",
  apiKey: process.env.SAIL_API_KEY,
});

const completion = await client.chat.completions.create({
  model: "zai-org/GLM-5.2-FP8",
  messages: [
    {
      role: "user",
      content: "What are the top 3 things to do in San Francisco?",
    },
  ],
});

console.log(completion.choices[0].message.content);

curl https://api.sailresearch.com/v1/chat/completions \
  -H "Authorization: Bearer $SAIL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zai-org/GLM-5.2-FP8",
    "messages": [
      {
        "role": "user",
        "content": "What are the top 3 things to do in San Francisco?"
      }
    ]
  }'
Run an AI modelRun leading open-source AI models with our OpenAI-compatible inference API.Create a SailboxGive long-horizon agents persistent compute that can run indefinitely.Send requests at scaleUse completion windows and background requests for large workloads.

​
Intelligence at scale

More agents thinking longer and harder, with space to act and explore, can do incredible things:
  • Detailuses Sail inference to deeply scan codebases for their most consequential yet hard-to-catch bugs
  • Jack & Jillruns large-scale deep research with Sail inference, matching job seekers’ resumes with job descriptions from thousands of employers
  • Wewon Browsecomp-Plus, the AI deep research benchmark, using open models running on Sail inference
  • Webuilt Redis in Rustwith a swarm of 4 long-horizon coding agents running on Sailboxes with Sail inference over 27 hours
Trust Center