Anthropic Claude

Configuring Anthropic Claude in Aimogen Pro, the available Claude models, context window behaviour, vision and tool-calling support, and common Claude errors.

Anthropic Claude models are widely used in Aimogen Pro for long-form writing and for agent work, where their long context windows and instruction-following matter.

Get a key#

  1. Sign in at console.anthropic.com.
  2. Go to Settings › API Keys and create a key.
  3. Add credit under Billing. Anthropic is prepaid; a zero balance returns a credit error immediately.

Configure#

Aimogen Pro › Settings › API Keys › Anthropic Claude API Keys

Paste one key per line. Select Test API key, then save.

Adding a Claude key makes every Claude model appear in every model selector across the plugin. You do not need to change Main API Service Provider Selector — Claude routing is decided by the model name.

Models#

Aimogen Pro 2.8.8 lists 17 Claude models.

ModelNotes
claude-opus-5Current top tier
claude-sonnet-5Balanced flagship, added in 2.8.4
claude-opus-4-8Added in 2.8.1
claude-opus-4-7, claude-opus-4-6, claude-sonnet-4-6Previous generation
claude-opus-4-5-20251101, claude-sonnet-4-5-20250929, claude-haiku-4-5-20251001Dated snapshots
claude-opus-4-1-20250805, claude-opus-4-20250514, claude-sonnet-4-20250514Claude 4
claude-3-7-sonnet-20250219, claude-3-5-haiku-20241022, claude-3-haiku-20240307Claude 3.x
claude-fable-5, claude-mythos-5Added in 2.8.2

The plugin tracks several capability sets internally so that features only offer models that support them: chat, 200K-context, and vision. Most current Claude models are in all three.

Capabilities in the plugin#

CapabilitySupported
Text generationyes
Chat and streamingyes
Vision (image input)yes, on current models
Tool calling / Chatbot Extensionsyes
AI Agentsyes
Embeddingsno — Anthropic does not offer an embeddings API. Use OpenAI, Google or Ollama
Image generationno — use an image provider
Assistants, Batch, fine-tuningno — these are OpenAI-only in the plugin

Context windows#

Claude models offer large context windows, and the plugin has a dedicated capability list for the 200K models. A large window is useful when you:

  • Feed long scraped pages into a prompt
  • Run retrieval with many chunks injected
  • Let an agent accumulate a long working history

It is not free: input tokens are billed. See Controlling cost and Tokens, context and output length.

Common problems#

invalid x-api-key or authentication_error The key is wrong, revoked, or has a stray character. Re-copy and use Test API key.

credit balance is too low Anthropic is prepaid. Top up in the console.

overloaded_error or HTTP 529 Anthropic is shedding load. This is transient. Reliability Mode classifies it as a provider failure and will cool the model down and fail over.

prompt is too long Input exceeds the window. Reduce injected context, shorten the source material, or use a larger-context model. See Tokens, context and output length.

Rate limits on a new account Anthropic tiers accounts by spend history. New accounts have low limits. Raise Delay Between API & Scraping Requests (ms) or add keys.

Endpoint used#

https://api.anthropic.com/v1/messages

Still stuck? Open a support ticket and include the diagnostics from Aimogen Pro › System & Logs › System Info.