static audit · 2026-10-25T22:30:47.251Z

langchain-ai/langchain

120 source files scanned · 9 LLM call site(s) · ~0% of LLM spend looks recoverable (static heuristics)

$105per 10k runs today (claude-sonnet-4-5, ~2000 prompt tokens)
$105per 10k runs with fixes applied

badge   ![TokenZero audit](https://tokenzero.sh/v1/badge/audit/aud_gHbIvB5Ssk0qf8LZY6KULw.svg)

repo badge   [![LLM spend audited by TokenZero](https://tokenzero.sh/v1/badge/repo/langchain-ai/langchain.svg)](https://tokenzero.sh/audit/langchain-ai/langchain)

Models referenced

claude-sonnet-4-5-20250929 ×1

Findings

No token-waste patterns detected in the scanned files.

Fix all of it with one line

OPENAI_BASE_URL=https://tokenzero.sh/v1
ANTHROPIC_BASE_URL=https://tokenzero.sh

Keys pass through untouched. Cache formatting and exact caching are on by default; cascade routing stays opt-in. How it works