danny-avila/LibreChat
120 source files scanned · 2 LLM call site(s) · ~8% of LLM spend looks recoverable (static heuristics)
$13.5per 10k runs today (gemini-2.5-flash, ~2000 prompt tokens)
$12.42per 10k runs with fixes applied
[](https://tokenzero.sh/audit/danny-avila/librechat)
Models referenced
gemini-2.5-flash-image ×1
Findings
| Severity | Where | Finding | Est. |
|---|---|---|---|
| medium | api/server/controllers/agents/openai.js:1039 | LLM call inside a loop Per-item calls repeat the full prompt prefix for every element. If items are independent and latency-tolerant, the provider Batch API is 50% cheaper, and OpenAI Flex processing (`service_tier: "flex"`) gives the same 50% on ordinary calls without restructuring; if the prefix is shared, caching pays off immediately. Fix: Batch where possible, or set `service_tier: "flex"` on OpenAI calls that can wait; otherwise keep the shared prefix cache-formatted (TokenZero does this automatically). | 25% |
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