API Docs — InferGate

1. Overview

InferGate provides a unified API gateway for approved users. The API is designed to be compatible with OpenAI-style chat completion requests, while allowing access to selected model routes through one endpoint.

Model availability, routing, pricing, and access limits may change. High-cost models may require manual approval or higher account limits.

2. Base URL

All API requests should use the following base URL:

Base URL

https://api.useinfergate.com/v1

3. Authentication

Pass your InferGate API key using the Authorization header.

Headers

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Keep API keys private. Do not expose keys in public repositories, client-side code, screenshots, or shared logs.

4. Available Models

The first commercial model set currently includes:

openai/gpt-5.5-pro
openai/gpt-5.5
openai/gpt-5.4-image-2
deepseek-v4-flash
deepseek-v4-pro

5. Pricing

InferGate pricing is usage-based and calculated per 1M tokens. The prices below are set at 1.5× the current official upstream reference price used for this model list.

Model ID	Model	Input Price	Output Price	Notes
`openai/gpt-5.5-pro`	OpenAI GPT-5.5 Pro	$45 / 1M tokens	$270 / 1M tokens	Premium reasoning model. Recommended for approved users and higher-value workloads.
`openai/gpt-5.5`	OpenAI GPT-5.5	$7.5 / 1M tokens	$45 / 1M tokens	Frontier general-purpose model for complex professional workloads.
`openai/gpt-5.4-image-2`	OpenAI GPT-5.4 Image 2	$12 / 1M tokens	$22.5 / 1M tokens	Multimodal model with image generation capabilities.
`deepseek-v4-flash`	DeepSeek V4 Flash	$0.14745 / 1M tokens	$0.2949 / 1M tokens	Cost-efficient model for fast responses and high-throughput workflows.
`deepseek-v4-pro`	DeepSeek V4 Pro	$0.6525 / 1M tokens	$1.305 / 1M tokens	Advanced reasoning model for coding, long-context analysis, and agent workflows.

Final billed usage may depend on request type, model route, tokenization, provider behavior, failed/retried requests, and account-specific settings.

Machine-readable price reference

pricing.json style reference

[
    {
      "model": "openai/gpt-5.5-pro",
      "input_price_per_1m_tokens": 45.0,
      "output_price_per_1m_tokens": 270.0
    },
    {
      "model": "openai/gpt-5.5",
      "input_price_per_1m_tokens": 7.5,
      "output_price_per_1m_tokens": 45.0
    },
    {
      "model": "openai/gpt-5.4-image-2",
      "input_price_per_1m_tokens": 12.0,
      "output_price_per_1m_tokens": 22.5
    },
    {
      "model": "deepseek-v4-flash",
      "input_price_per_1m_tokens": 0.14745,
      "output_price_per_1m_tokens": 0.2949
    },
    {
      "model": "deepseek-v4-pro",
      "input_price_per_1m_tokens": 0.6525,
      "output_price_per_1m_tokens": 1.305
    }
]

6. Chat Completions

Endpoint:

POST /chat/completions

curl https://api.useinfergate.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5.5",
    "messages": [
      {
        "role": "user",
        "content": "Hello from InferGate."
      }
    ]
  }'

PowerShell example

Windows PowerShell

$headers = @{
  "Authorization" = "Bearer YOUR_API_KEY"
  "Content-Type"  = "application/json"
}

$body = @{
  model = "openai/gpt-5.5"
  messages = @(
    @{
      role = "user"
      content = "Hello from InferGate."
    }
  )
} | ConvertTo-Json -Depth 10

Invoke-RestMethod `
  -Uri "https://api.useinfergate.com/v1/chat/completions" `
  -Method Post `
  -Headers $headers `
  -Body $body

7. Error Format

Example error

{
  "error": {
    "type": "invalid_request_error",
    "message": "Missing or invalid API key.",
    "code": "invalid_api_key"
  }
}

Status Code	Meaning
400	Invalid request payload.
401	Missing or invalid API key.
403	Model, group, or account access not allowed.
429	Rate limit, quota, or balance limit exceeded.
500	Internal gateway error.
503	Upstream provider or route temporarily unavailable.

8. Limits and Usage Controls

Accounts may have request limits, spend limits, model limits, group limits, and abuse-prevention checks. Access to premium models may require manual review.

9. Request Access

To request an account, commercial access, or higher usage limits, submit the early-access form.

Request Early Access

InferGate API Docs