Use the OpenAI Python SDK with Eden AI

Set api_key and base_url on an existing OpenAI client and switch model ids to the provider/model format; your client.chat.completions calls and their response objects stay the same.

Overview

Eden AI’s V3 chat completions endpoint is OpenAI-compatible. Models from OpenAI, Anthropic, Google, Cohere and other providers use the same call. The other OpenAI APIs (Files, Assistants, Batches) are not served under /v3; uploads go through /v3/upload. For streaming, see Streaming.

Installation

pip install openai

poetry add openai

Quick Start

Configure the OpenAI client to use Eden AI:

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",  # Get from https://app.edenai.run
    base_url="https://api.edenai.run/v3"
)

response = client.chat.completions.create(
    model="openai/gpt-latest",
    messages=[
        {"role": "user", "content": "Hello! How are you?"}
    ]
)

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

Model naming

Models use the provider/model format, for example openai/gpt-latest, anthropic/claude-sonnet-latest and vertex/gemini-flash-latest. Browse the catalog in List LLM Models.

Multi-Turn Conversations

Build conversational applications with message history:

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.edenai.run/v3"
)

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What is the capital of France?"}
]

response = client.chat.completions.create(
    model="vertex/gemini-flash-latest",
    messages=messages
)

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

# Add assistant response to history
messages.append({"role": "assistant", "content": assistant_response})

# Continue conversation
messages.append({"role": "user", "content": "What's the population?"})

response = client.chat.completions.create(
    model="vertex/gemini-flash-latest",
    messages=messages
)

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

Vision Capabilities

Send images to a model that accepts them: its capabilities.input_modalities in List LLM Models includes image.

from openai import OpenAI
import requests

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.edenai.run/v3"
)

# First, upload the image to get a file_id
with open("image.jpg", "rb") as f:
    upload_response = requests.post(
        "https://api.edenai.run/v3/upload",
        headers={"Authorization": "Bearer YOUR_API_KEY"},
        files={"file": f}
    )
file_id = upload_response.json()["file_id"]

# Use the file_id in a chat message
response = client.chat.completions.create(
    model="openai/gpt-latest",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {"type": "file", "file": {"file_id": file_id}}
            ]
        }
    ]
)

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

Error Handling

The SDK raises a subclass of openai.APIStatusError for most statuses; 402 and 451 have none. These are the statuses specific to Eden AI:

Status	Exception	Meaning
401	`openai.AuthenticationError`	The key is not accepted. The detail is `Invalid API key` for a wrong or revoked `sk-eden-...` key, `Custom token has expired.` for an expired one, and `Invalid token` for other token shapes.
402	`openai.APIStatusError`	Insufficient credit. The SDK has no subclass for 402.
403	`openai.PermissionDeniedError`	A compliance or guardrail rule refused the request.
429	`openai.RateLimitError`	Rate limit exceeded. The SDK retries twice before raising; the `Retry-After` header says how long to wait.
451	`openai.APIStatusError`	No provider of the model runs in a region that the endpoint you called, or your organization’s compliance policy, allows, or, for a routing policy, every target was dropped for its region.

An unknown model id answers 400 Model(s) not found or inactive, an invalid request 422, and a provider failure 5xx. The APIStatusError handler below catches them with the others.

from openai import OpenAI
import openai

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.edenai.run/v3"
)

try:
    response = client.chat.completions.create(
        model="openai/gpt-latest",
        messages=[{"role": "user", "content": "Hello!"}]
    )

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

except openai.AuthenticationError as e:
    print(f"Authentication failed: {e}")
except openai.PermissionDeniedError as e:
    print(f"Request refused: {e}")
except openai.RateLimitError as e:
    print(f"Rate limit exceeded: {e}")
except openai.APIStatusError as e:
    # 402 (insufficient credit), 451 (region refusal) and any other status
    print(f"API error {e.status_code}: {e}")
except openai.APIConnectionError as e:
    print(f"Could not reach the API: {e}")

Environment Variables

EDEN_AI_API_KEY=your_api_key_here

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("EDEN_AI_API_KEY"),
    base_url="https://api.edenai.run/v3"
)

Next Steps

Chat Completions - Core LLM endpoint
Streaming - Stream tokens as they are generated
File Attachments - Working with images and documents
List LLM Models - Browse available providers and models

V3 Documentation

Quick Start

Overview

LLMs

Expert Models

General

Organization

Data Governance

Integrations

Use the OpenAI Python SDK with Eden AI

Overview

Installation

Quick Start

Model naming

Multi-Turn Conversations

Vision Capabilities

Error Handling

Environment Variables

Next Steps

​Overview

​Installation

​Quick Start

​Model naming

​Multi-Turn Conversations

​Vision Capabilities

​Error Handling

​Environment Variables

​Next Steps

Overview

Installation

Quick Start

Model naming

Multi-Turn Conversations

Vision Capabilities

Error Handling

Environment Variables

Next Steps