Skip to main content

​
Retries

Certain errors are automatically retried 2 times by default, with a short exponential backoff. The following errors are retried automatically:
  • Connection errors (network connectivity problems)
  • 408 Request Timeout
  • 409 Conflict
  • 429 Rate Limit
  • 500+ Internal Server errors

​
Configuring retries globally

You can configure retry settings when creating the client: 
from dedalus_labs import Dedalus

# Disable retries
client = Dedalus(
    max_retries=0,
)

# Increase retries
client = Dedalus(
    max_retries=5,
)
The default is max_retries=2.

​
Configuring retries per-request

You can also configure retries on a per-request basis using with_options(): 
from dedalus_labs import Dedalus

client = Dedalus()

# This request will retry up to 5 times
client.with_options(max_retries=5).chat.completions.create(
    model="openai/gpt-5-nano",
    messages=[
        {
            "role": "system",
            "content": "You are Stephen Dedalus. Respond in morose Joycean malaise.",
        },
        {
            "role": "user",
            "content": "Hello, how are you today?",
        },
    ],
)

​
Timeouts

By default, requests time out after 1 minute (60 seconds). You can configure this with the timeout option.

​
Basic timeout configuration

Set a simple timeout in seconds: 
from dedalus_labs import Dedalus

# Configure the default for all requests (20 seconds)
client = Dedalus(
    timeout=20.0,
)

​
Granular timeout control

For more granular control, use httpx.Timeout to configure different timeout values for different phases of the request: 
import httpx
from dedalus_labs import Dedalus

client = Dedalus(
    timeout=httpx.Timeout(
        60.0,      # Total timeout
        read=5.0,   # Time to read response
        write=10.0, # Time to send request
        connect=2.0 # Time to establish connection
    ),
)

​
Per-request timeouts

Override the timeout for specific requests: 
from dedalus_labs import Dedalus

client = Dedalus()

# This request will timeout after 5 seconds
client.with_options(timeout=5.0).chat.completions.create(
    model="openai/gpt-5-nano",
    messages=[
        {
            "role": "system",
            "content": "You are Stephen Dedalus. Respond in morose Joycean malaise.",
        },
        {
            "role": "user",
            "content": "Hello, how are you today?",
        },
    ],
)

​
Handling timeout errors

When a request times out, an APITimeoutError is thrown: 
import dedalus_labs
from dedalus_labs import Dedalus

client = Dedalus(timeout=5.0)

try:
    response = client.chat.completions.create(
        model="openai/gpt-5-nano",
        messages=[{"role": "user", "content": "Hello"}],
    )
except dedalus_labs.APITimeoutError as e:
    print(f"Request timed out: {e}")
Note that requests that time out are retried twice by default unless you configure max_retries=0.

Build docs developers (and LLMs) love

Get started for freeTalk to us