Loading page
About this tool
Compare real-time and batch LLM pricing for bulk jobs and see the saving, blended cost and latency you trade for it.
Batch LLM endpoints sell the same tokens at a discount in exchange for asynchronous delivery: the discount applies to input and output alike, so batch cost = real-time cost × (1 − discount). This calculator prices a bulk job both ways, then blends them when only part of the workload can tolerate the wait — blended = real-time requests × per-request cost + batched requests × discounted cost. The published discount on the OpenAI Batch API, the Anthropic Message Batches API and Gemini batch mode is 50%, with a completion window of about 24 hours.
Open Batch API Savings Calculator on AltFTool — it loads instantly in your browser.
Enter the values you already know.
Fine-tune the options to match your scenario.
Read the result and use it in your planning or reporting.
Prices a mixed workload instead of assuming everything can be made asynchronous.
Reports the discounted per-million input and output rates you actually pay.
Multiplies one job by its monthly cadence for a run-rate and annual figure.
50% on the major providers — OpenAI, Anthropic and Google all publish a half-price batch tier that applies to both input and output tokens. A job costing $9,000 synchronously costs $4,500 if the whole thing is batched.
The published target is completion within 24 hours of submission, and many jobs finish far sooner. There is no latency guarantee below that window, which is why batch suits offline work — enrichment, classification, evaluation, backfills — and not anything a user is waiting on.
The saving scales linearly with the batchable share. At a 50% discount, moving 80% of a workload to batch cuts the bill by 40%, and moving 50% cuts it by 25%. Weigh that against the one-off cost of building job submission, polling and partial-failure handling.
Not usefully in most cases. Batch jobs are queued and executed on the provider's schedule, so a cache written for one request is unlikely to still be warm for the next, and vendors generally do not apply cache-read pricing inside batch runs. Model the two savings separately rather than multiplying them.
Add the Batch API Savings Calculatorwidget to your blog or website — free, responsive, no signup. Just keep the “Widget by AltFTool” credit link visible.
<iframe src="https://www.altftool.com/embed/widget/batch-api-savings-calculator"
title="Batch API Savings Calculator — free AltFTool widget"
width="100%" height="640" style="border:0;border-radius:12px;overflow:hidden"
loading="lazy" referrerpolicy="no-referrer-when-downgrade"></iframe>
<p style="font-size:12px;margin:4px 0 0">Widget by <a href="https://www.altftool.com/tools/all/batch-api-savings-calculator?utm_source=embed&utm_medium=widget">AltFTool — free online tools</a></p>