Skip to main content
Watch: I Built an AI App Without Leaving Databricks

AI API error index - verified October 18, 2026

AI API errors, explained and fixed

Look up the status code and error type your API call returned, then read what it means, why it happens, the documented fix and whether retrying helps. 204 errors from 12 model APIs, copied from each provider's official error docs.

Pick your provider

Common errors across providers

The same HTTP status means different things at different APIs. Each row lists the providers whose official docs define an error with that status; the count is how many distinct errors they document for it.

413Content Too Large

Not in the docs: Amazon Bedrock, Azure OpenAI, Cohere API, DeepSeek API, Gemini API, Mistral API, OpenAI API, Vertex AI, xAI API

422Unprocessable Content

Not in the docs: Amazon Bedrock, Anthropic API, Azure OpenAI, Cohere API, Gemini API, Vertex AI

529Non-standard

Not in the docs: Azure OpenAI, Cohere API, DeepSeek API, Gemini API, Groq API, Mistral API, OpenAI API, OpenRouter, Vertex AI, xAI API

Method

How we verify

Every error, cause and fix is copied from the provider's official error documentation or its official SDK repository, then re-read by an independent reviewer before publishing. The retry badge says only what those docs say: when they do not address retrying, the page says so.

Retry? Yes
The official docs treat this error as retryable.
Retry? No
The official docs treat this error as not retryable: fix the cause before sending the request again.
Retry? Not stated
The official docs do not say whether this error is retryable.

Newsletter

Know when the APIs change

One email per week. New model releases, API changes, what broke, and how developers are working around it.

Debug the tool, then check the limits

Coding agent errors

Verbatim messages from Claude Code, Codex, Cursor and more, each with the documented fix.

Rate limits

Per-model request and token limits for each API, copied from the official docs.

Model registry

First-party API prices, context windows and weights for each model.