# Use Batch API for AI agent

**URL:** <https://community.n8n.io/t/use-batch-api-for-ai-agent/227071>\
**Category:** Questions\
**Tags:** workflow-building\
**Created:** [November 26, 2025, 6:19pm UTC](https://community.n8n.io/t/use-batch-api-for-ai-agent/227071 "2025-11-26T18:19:12Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Kent\_Lee](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/kent_lee/32/62479_2.png) [@Kent\_Lee](https://community.n8n.io/u/Kent_Lee)\
**Post date:** [November 26, 2025, 6:19pm UTC](https://community.n8n.io/t/use-batch-api-for-ai-agent/227071/1 "2025-11-26T18:19:12Z")

</div>

Can we use Batch API for AI agent? How to set it?

I am using Claude Sonnet 4

## Information on your n8n setup

- **n8n version: latest**
- **Running n8n via n8n cloud**

---

<div class="post-metadata">

**Author:** ![NCione](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/ncione/32/63903_2.png) [@NCione](https://community.n8n.io/u/NCione)\
**Post date:** [November 26, 2025, 6:42pm UTC](https://community.n8n.io/t/use-batch-api-for-ai-agent/227071/2 "2025-11-26T18:42:07Z")

</div>

Hi @Kent_Lee, in the Add Options section of the AI Agent theres a Batch Processing Option which you could use. If you’re looking for Batching through the API itself, I don’t see any native features within Anthropic’s chat model connector, but you could use an HTTP request node and write the payload yourself in line with their API docs. Hope this helps!

---

<div class="post-metadata">

**Author:** ![Kent\_Lee](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/kent_lee/32/62479_2.png) [@Kent\_Lee](https://community.n8n.io/u/Kent_Lee)\
**Post date:** [November 26, 2025, 7:41pm UTC](https://community.n8n.io/t/use-batch-api-for-ai-agent/227071/3 "2025-11-26T19:41:22Z")

</div>

Is this batch processing option mean calling the Batch API or just the node itself batching the calls?

---

<div class="post-metadata">

**Author:** ![NCione](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/ncione/32/63903_2.png) [@NCione](https://community.n8n.io/u/NCione)\
**Post date:** [November 26, 2025, 7:50pm UTC](https://community.n8n.io/t/use-batch-api-for-ai-agent/227071/4 "2025-11-26T19:50:47Z")

</div>

This instance is the node itself batching calls (to combat rate limiting). I’m not sure if any of the “message a model” nodes/tools in n8n natively support batching API endpoints. Your best bet would be to manually hit the batching API endpoint with an HTTP request node in my opinion

---

<div class="post-metadata">

**Author:** ![system](https://dub1.discourse-cdn.com/n8n/user_avatar/community.n8n.io/system/32/36532_2.png) [@system](https://community.n8n.io/u/system)\
**Post date:** [February 24, 2026, 7:51pm UTC](https://community.n8n.io/t/use-batch-api-for-ai-agent/227071/5 "2026-02-24T19:51:42Z")

</div>

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.