Exceptions

Error types raised by the MinT API

Base Exceptions

TinkerError

Root exception for all MinT errors.

All MinT-specific exceptions inherit from this class.

APIError

Base class for API-related errors.

Properties:

body - Response body containing error details

Connection Errors

APIConnectionError

Raised when connection to the MinT server fails.

APITimeoutError

Raised when a request times out.

Inherits from: APIConnectionError

APIResponseValidationError

Raised when server response doesn’t match expected schema.

HTTP Status Errors

Exception	Status Code	Description
`BadRequestError`	400	Invalid request format or parameters
`AuthenticationError`	401	Missing or invalid credentials
`PermissionDeniedError`	403	Insufficient permissions for operation
`NotFoundError`	404	Requested resource doesn’t exist
`ConflictError`	409	Resource state conflict
`UnprocessableEntityError`	422	Request is semantically invalid
`RateLimitError`	429	Too many requests (rate limit exceeded)
`InternalServerError`	500+	Server-side error

Async Errors

RequestFailedError

Raised when an async request fails to complete successfully. This can occur due to:

Transient server issues
Resource contention
Session timeout (see Limits and Quotas)

Properties:

request_id - Unique identifier for debugging (provide this when contacting support)

Retry Strategy:

import asyncio
from mint import RequestFailedError
 
async def sample_with_retry(sampling_client, prompt, params, max_retries=3):
    for attempt in range(max_retries):
        try:
            return await sampling_client.sample_async(prompt=prompt, sampling_params=params)
        except RequestFailedError as e:
            if attempt == max_retries - 1:
                raise
            wait_time = 2 ** attempt  # 1s, 2s, 4s
            print(f"Request failed (id={e.request_id}), retrying in {wait_time}s...")
            await asyncio.sleep(wait_time)

If errors persist after retries, create a fresh sampling client:

sampling_client = training_client.save_weights_and_get_sampling_client()

Usage Example

import mint
from mint.exceptions import (
    AuthenticationError,
    NotFoundError,
    RateLimitError
)
 
try:
    service_client = mint.ServiceClient()
    training_client = service_client.create_lora_training_client(
        base_model="Qwen/Qwen3-4B-Instruct-2507"
    )
except AuthenticationError:
    print("Invalid API key")
except NotFoundError:
    print("Model not found")
except RateLimitError:
    print("Rate limit exceeded, please wait")
except mint.TinkerError as e:
    print(f"MinT error: {e}")

Parameters Quickstart