Your prompt is processed in memory only — never written to disk, never sent to a model for inference. We log content-free metadata for abuse prevention. View audit policy →
Token breakdown
| Model | Tokenizer | Tokens | $/request | $/1M requests |
|---|
Input pricing only. We do not estimate output cost because output length depends on the model and prompt. Pricing effective 2026-04-01.
Structural composition
What we noticed
Estimated savings
Matched cohort:
This is a directional estimate from cohort matching against Ordica's blind-judged benchmark, not a guarantee for this specific prompt. See the methodology →
How this works
- Real tokenizers, not approximations. GPT-4o uses tiktoken cl100k_base locally. Claude uses Anthropic's count_tokens API. Gemini uses Google's countTokens REST endpoint. Grok uses cl100k_base as an approximation (xAI does not publish a tokenizer).
- Section parsing is heuristic. We detect system instructions, few-shot examples, RAG context, tool definitions, and user queries from public structural markers (role-based JSON, "Document N:", "Example:", etc.). Confidence is reported per analysis.
- Savings estimates come from a benchmark, not your prompt. We match your prompt's structure to a cohort in our 200-prompt × 4-provider blind-judged benchmark and return that cohort's percentile range. We never run our compression engine on your input.
- Compression preserves quality better in some categories than others. We surface per-provider equivalence scores (1-5 scale) so you can see where compression is reliable and where it isn't. Don't trust this page — verify the cohort table yourself.
Privacy contract
- What we log: timestamp, salted+hashed IP (rotates daily), token count, matched cohort, parser confidence, response duration.
- What we never log: your prompt, any section text, your identity, raw IPs, cookies, session tokens.
- What leaves this server: Anthropic and Gemini token counting send your prompt to those vendors for tokenization only. Their data policies apply. OpenAI and Grok counts are computed entirely on this server.
- Verification: the audit endpoint shows the policy and the last 20 (content-free) entries.
Public cohort data → Full methodology → Audit log →
Legal: Terms of Service · Privacy Policy · Data Processing Agreement (request a signed copy at legal@ordica.ai)