Free LLM FinOps tools
Every tool runs on the pricing table the gateway signs receipts with and on its own tokenizer; results render on the server so curl sees the same numbers as a browser.
7 tools, no sign-up: on the default scenario the cost calculator shows gpt-6.1-sol going from $700 to $232 a month (66.88%) with the prefix cached, 10% exact repeats and 40% of prompts cascaded, all from pricing table 2026-10-26; the latency tool shows the measured 1.52 ms p50 gateway overhead and this gateway's live figure.
· Source: pricing table 2026-10-26, gateway tokenizer, scripts/bench result files
At a glance
- Tools
- 7
- Models priced
- 62
- Tokenizers
- openai, anthropic, gemini, xai, deepseek
- Default-scenario saving (gpt-6.1-sol)
- 66.88% (computed)
- Measured overhead
- 1.52 ms p50
- Server-rendered
- yes; interactive with JavaScript
- Data stored
- none; token counts are computed and discarded
- Price
- free
The tools
| Tool | What it does |
|---|---|
| LLM cost calculator | Price any workload on any of the catalogue models, then see what caching, exact repeats and cascade routing take off. |
| Prompt caching savings estimator | Paste a system prompt: exact token count, whether it clears the cache minimum, break-even reads and monthly dollars saved. |
| Cascade routing simulator | See what routing a share of simple prompts to the cheaper sibling saves, and how the parity breaker protects quality. |
| LLM tokenizer and compressor | Count tokens for five providers and preview the lossless whitespace compaction the gateway applies to tool schemas. |
| GitHub repo token-waste auditor | Audit any public repository for uncached prompts and expensive or retired model ids; get a shareable receipt page and badge. |
| Gateway latency and TTFB benchmark | Measured gateway overhead per path, streaming first byte and the LiteLLM head-to-head, plus the live figure from this gateway. |
| OpenAI to Claude cost converter | Map an OpenAI model to the Claude model of the same tier (and back) and price the same workload on both. |
All tools
LLM cost calculatorPrompt caching savings estimatorCascade routing simulatorLLM tokenizer and compressorGitHub repo token-waste auditorGateway latency and TTFB benchmarkOpenAI to Claude cost converter
Questions people ask
Do the tools need an account?
No. They run on the public pricing table and tokenizer; a workspace is only needed to route traffic and collect receipts.
Are the numbers the same the gateway bills with?
Yes: the same pricing table (2026-10-26) and the same tokenizer price every signed receipt, so a tool estimate and a receipt can be compared line by line.
Where do the latency numbers come from?
From the benchmark scripts published with their raw JSON on /benchmarks, and live from /v1/status for this gateway.