Documentation

frogAPI provides a unified gateway to the world's leading AI models through a single, OpenAI-compatible REST API. Change one line of code in your existing application to route requests through frogAPI and instantly benefit from wholesale pricing.

All endpoints are fully compatible with the OpenAI SDK, meaning you can use the official openai Python and Node.js libraries with zero modifications beyond updating the base URL and API key.

Authentication

All API requests require an API key passed in the Authorization header using the Bearer scheme. API keys are prefixed with frog_sk_ for easy identification.

You can generate and manage your API keys from the Dashboard. Each account receives one API key by default.

http

GET /v1/chat/completions HTTP/1.1
Host: frogapi.app
Authorization: Bearer frog_sk_your_api_key_here
Content-Type: application/json

Keep your API key secure

Never expose your API key in client-side code, public repositories, or browser-accessible locations. Use environment variables and server-side proxying.

Quick Start

Get up and running with frogAPI in under 60 seconds. Since frogAPI is fully OpenAI-compatible, integration is as simple as changing your base URL.

1Install the OpenAI SDK

bash

pip install openai

2Make your first request

python

from openai import OpenAI

client = OpenAI(
    api_key="frog_sk_your_api_key_here",
    base_url="https://frogapi.app/v1"
)

response = client.chat.completions.create(
    model="gpt-5.2",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello, frogAPI!"}
    ]
)

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

3Enable streaming (optional)

Stream responses in real-time using Server-Sent Events. Simply set stream: true in your request.

python

from openai import OpenAI

client = OpenAI(
    api_key="frog_sk_your_api_key_here",
    base_url="https://frogapi.app/v1"
)

stream = client.chat.completions.create(
    model="gpt-5.2",
    messages=[{"role": "user", "content": "Tell me a joke"}],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Endpoints

POST/v1/chat/completions

Creates a model response for the given chat conversation. This is the primary endpoint for all chat-based AI interactions.

Request Body

Parameter	Type	Required	Description
`model`	string	Yes	ID of the model to use (e.g. gpt-5.2)
`messages`	array	Yes	List of messages in the conversation
`temperature`	number	No	Sampling temperature, 0–2. Default: 1
`max_tokens`	integer	No	Maximum tokens to generate
`stream`	boolean	No	Enable Server-Sent Events streaming
`top_p`	number	No	Nucleus sampling parameter. Default: 1

Example Response

json

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1709000000,
  "model": "gpt-5.2",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 9,
    "total_tokens": 29
  }
}

GET/v1/models

Lists all available models currently served by frogAPI. Returns an array of model objects with metadata.

curl https://frogapi.app/v1/models \
  -H "Authorization: Bearer frog_sk_your_api_key_here"

Models

frogAPI serves the following models. Use the exact model ID in your API requests. All models are available at the same source-matching pricing — see the Pricing page for details.

Model ID	Provider	Context	Best For
gpt-5.4-mini	OpenAI	128K	High-performance compact
gpt-5.4-nano	OpenAI	128K	Fast & efficient
gpt-5.3	OpenAI	128K	Latest flagship
gpt-5.2	OpenAI	128K	Flagship
gpt-5-mini	OpenAI	128K	Balanced performance & cost
gpt-5-nano	OpenAI	128K	Ultra-fast, low-latency tasks
gemini-3.1-pro	Google	1M	Latest flagship
gemini-3-flash	Google	1M	Fast balanced
gemini-3.1-flash-lite	Google	1M	Ultra-fast value
gemini-2.5-pro	Google	1M	Advanced reasoning
kimi-k2.5	Kimi	128K	Advanced reasoning
grok-4.1-fast	xAI	128K	Fast reasoning
grok-4.1-fast-nr	xAI	128K	Fast non-reasoning
mistral-large-3	Mistral	128K	Multilingual & code generation
deepseek-v3.2	DeepSeek	64K	Cost-effective reasoning

Note

All models accept the same OpenAI-compatible request format. To switch models, simply change the model parameter — no other code changes needed.

Gemini Models

frogAPI currently serves all Gemini models below through the same OpenAI-compatible /v1/chat/completions endpoint.

Gemini Model	Best For	Input / 1M	Output / 1M
gemini-3.1-pro	Highest-quality Gemini reasoning	$2.00	$12.00
gemini-3-flash	Lower-latency general-purpose generation	$0.50	$3.00
gemini-3.1-flash-lite	Cheapest Gemini tier for lightweight tasks	$0.25	$1.50
gemini-2.5-pro	Advanced reasoning on Gemini 2.5	$1.25	$10.00

Compatibility

Gemini models use the same messages format as every other frogAPI model.

max_tokens and max_completion_tokens are both accepted.

stop is supported and mapped to Gemini stop sequences.

Gemini-specific Notes

frequency_penalty and presence_penalty are ignored for Gemini requests.

Base64 image data URLs are supported in message content arrays.

Remote image URLs are passed through as text references rather than uploaded binaries.

Error Handling

frogAPI uses standard HTTP status codes. Errors return a JSON body with a descriptive error object containing a human-readable message, error type, and machine-readable code.

json

{
  "error": {
    "message": "Invalid API key provided.",
    "type": "authentication_error",
    "code": "invalid_api_key"
  }
}

Status	Type	Description
400	invalid_request	Your request is malformed or missing required fields
401	authentication_error	Invalid or missing API key
402	insufficient_funds	Your account balance is too low
404	not_found	The requested model or resource does not exist
429	rate_limit_exceeded	You have exceeded your rate limit
500	server_error	An unexpected error occurred on our end
503	service_unavailable	The upstream provider is temporarily unavailable

Retry Strategy

For 429 and 5xx errors, implement exponential backoff with jitter. Respect the Retry-After header when present. A good starting point is 1 second, doubling on each retry up to a maximum of 30 seconds.

Rate Limits

Limits are applied per user, per model.

RPM

Requests/min

TPM

Tokens/min

RPD

Requests/day

When Exceeded

API returns 429. Check Retry-After.

Daily Reset

Daily limit resets at 00:00 UTC.

For detailed per-model limits, see the Rate Limits page.

SDKs & Libraries

frogAPI is fully OpenAI-compatible. Any library or framework that works with the OpenAI API works with frogAPI — just update the base URL and API key.

Python

openai

Install

pip install openai

Base URL

base_url="https://frogapi.app/v1"

Node.js

openai

Install

npm install openai

Base URL

baseURL: "https://frogapi.app/v1"

LangChain

langchain-openai

Install

pip install langchain-openai

Base URL

openai_api_base="https://frogapi.app/v1"

LiteLLM

litellm

Install

pip install litellm

Base URL

api_base="https://frogapi.app/v1"

Migration Tip

If you already use OpenAI SDKs, only change base_url and API key.