Secure OpenAI Inference with DeepFellow

Using OpenAI models through DeepFellow is safer than using them directly thanks to our anonymization layer. DeepFellow provides a seamless, OpenAI-compatible API that makes it easy to integrate advanced language models with your existing applications.

This page covers the Chat Completions API (/v1/chat/completions). Some OpenAI models, such as gpt-5.1-codex and o1-pro, only support the OpenAI Responses API (/v1/responses) and will return an error on this endpoint; see OpenAI Responses API with DeepFellow. Check the "Models" list on the openai service in Infra Web Panel for all currently available models.

You need to have an OpenAI API Key and "openai" service installed - read how to install services in Infra Web Panel guide.

curl -X 'POST' \
  'https://deepfellow-server-host/v1/chat/completions' \
  -H 'Authorization: Bearer DEEPFELLOW-PROJECT-API-KEY' \
  -H 'Content-Type: application/json' \
  -d '{
  "messages": [
    {
      "content": "You are a helpful assistant.",
      "role": "system"
    },
    {
      "content": "Explain how AI works.",
      "role": "user"
    }
  ],
  "model": "gpt-4.1-mini"
}'

from openai import OpenAI

client = OpenAI(
    api_key="DEEPFELLOW-PROJECT-API-KEY",
    base_url="https://deepfellow-server-host/v1",
)

response = client.chat.completions.create(
    model="gpt-4.1-mini",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {
            "role": "user",
            "content": "Explain how AI works."
        }
    ]
)

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

import OpenAI from 'openai';

const openai = new OpenAI({
    apiKey: 'DEEPFELLOW-PROJECT-API-KEY',
    baseURL: 'https://deepfellow-server-host/v1'
});

const response = await openai.chat.completions.create({
    model: 'gpt-4.1-mini',
    messages: [
        { role: 'system', content: 'You are a helpful assistant.' },
        {
            role: 'user',
            content: 'Explain how AI works.'
        }
    ]
});

console.log(response.choices[0].message);