Use AIUsage with LlamaIndex

One endpoint change. Under a minute. Same prompts, same outputs, 70–90% smaller Claude bill.

Setup

Get your AIUsage endpoint URL

Point LlamaIndex at it

Paste this into your setup:

from llama_index.llms.anthropic import Anthropic

llm = Anthropic(
    model="claude-3-5-sonnet-latest",
    base_url="https://aiusage.ai/claude",
)
# Your existing index + query pipelines work unchanged.

Use LlamaIndex normally

RAG queries, structured output, and agent abstractions all pass through.

Your Anthropic API key stays yours. AIUsage forwards it transparently to Anthropic. It is never stored, never logged. You can delete your AIUsage account any time and your integration reverts to vanilla Claude in one env-var change.

What this gets you on LlamaIndex

Every LlamaIndex request that used to hit Anthropic at retail price now runs through AIUsage's audit layer first. On the six workloads we've verified, the measured delta was 76–84% on the same prompts. Your LlamaIndex behavior — autocomplete, chat, agent mode, whatever — doesn't change.

Audit before you commit

The audit tool is free and shows you exactly what you would have paid. No signup required for the number. If your expected delta is under 40%, the tool tells you outright and you walk. Run the audit →

Audit my own LlamaIndex usage →