From 10c88358e1e5b445cf5f3280834561c0e9231d5c Mon Sep 17 00:00:00 2001
From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com>
Date: Mon, 31 Aug 2026 13:13:57 +0000
Subject: [PATCH 1/2] docs: clarify webhook payload, FlashBoot behavior, skills
requirements, and pricing sources
---
get-started/agent-skills.mdx | 4 ++++
pods/pricing.mdx | 4 +++-
serverless/endpoints/endpoint-configurations.mdx | 4 ++++
serverless/endpoints/send-requests.mdx | 16 +++++++++++++++-
serverless/pricing.mdx | 2 ++
5 files changed, 28 insertions(+), 2 deletions(-)
diff --git a/get-started/agent-skills.mdx b/get-started/agent-skills.mdx
index d7360c890..53131a239 100644
--- a/get-started/agent-skills.mdx
+++ b/get-started/agent-skills.mdx
@@ -9,6 +9,10 @@ The Runpod skills plugin teaches your coding agent how to run GPU workloads on R
It works with [Claude Code](https://code.claude.com/docs), [Codex](https://github.com/openai/codex), [Cursor](https://cursor.com/), [GitHub Copilot](https://github.com/features/copilot), [Windsurf](https://codeium.com/windsurf), [Cline](https://github.com/cline/cline), and [many other AI agents](https://www.skills.sh/).
+
+The skills plugin requires a coding agent with terminal access and Node.js installed (the install command uses `npx`). It can't be installed into chat apps like the ChatGPT or Claude mobile apps, which don't provide a shell.
+
+
## Quick start
Install the skills with a single command that works with every agent:
diff --git a/pods/pricing.mdx b/pods/pricing.mdx
index d29ab4a74..21b924944 100644
--- a/pods/pricing.mdx
+++ b/pods/pricing.mdx
@@ -13,7 +13,9 @@ import { MachineTooltip } from "/snippets/tooltips.jsx";
Runpod offers custom pricing plans for large scale and enterprise workloads. [Contact our sales team](https://ecykq.share.hsforms.com/2MZdZATC3Rb62Dgci7knjbA) to learn more.
-Pods are billed by the second for compute and storage, with no fees for data ingress or egress. Find the latest GPU pricing on the [Runpod console](https://www.console.runpod.io/pods) during Pod deployment.
+Pods are billed by the second for compute and storage, with no fees for data ingress or egress.
+
+Per-GPU hourly rates and savings plan discounts aren't listed in the documentation because they vary by GPU type and cloud type (Secure or Community) and change over time. Find current rates on the [Runpod pricing page](https://www.runpod.io/pricing) or in the [Runpod console](https://www.console.runpod.io/pods) during Pod deployment.
## Pricing options
diff --git a/serverless/endpoints/endpoint-configurations.mdx b/serverless/endpoints/endpoint-configurations.mdx
index 540cd6b01..957b50938 100644
--- a/serverless/endpoints/endpoint-configurations.mdx
+++ b/serverless/endpoints/endpoint-configurations.mdx
@@ -121,6 +121,10 @@ Results are permanently deleted after retention expires.
Reduces cold starts by retaining worker state after spin-down, allowing faster "revival" than fresh boots. Most effective on endpoints with consistent traffic where workers frequently cycle between active and idle.
Both new GPU and CPU endpoints will have FlashBoot enabled by default, and you can edit existing endpoints to enable or disable FlashBoot.
+State retention is opportunistic, not guaranteed. If a worker's retained state is no longer available, for example after a long period without traffic, the worker starts with a standard cold start instead. To guarantee low latency, set [active workers](#active-workers) above zero.
+
+When managing endpoints through the [REST API](/api-reference-v2/serverless/create-a-serverless-endpoint), the `flashboot` field accepts three modes: `OFF`, `FLASHBOOT`, and `PRIORITY_FLASHBOOT` (FlashBoot with priority capacity).
+
### Model
Select from [cached models](/serverless/endpoints/model-caching) to schedule workers on with model files pre-loaded. Significantly reduces model loading time during initialization.
diff --git a/serverless/endpoints/send-requests.mdx b/serverless/endpoints/send-requests.mdx
index 7ff04208b..f574ce04b 100644
--- a/serverless/endpoints/send-requests.mdx
+++ b/serverless/endpoints/send-requests.mdx
@@ -94,7 +94,21 @@ Receive notifications when jobs complete by specifying a webhook URL:
}
```
-Your webhook should return a `200` status code. If the call fails, Runpod retries up to 2 more times with a 10-second delay.
+When your job completes, Runpod sends a `POST` request to your webhook URL. The request body contains the same information as the [`/status` operation](/serverless/endpoints/operation-reference#status), including the job `id`, `status`, and `output`:
+
+```json
+{
+ "delayTime": 31618,
+ "executionTime": 1437,
+ "id": "60902e6c-08a1-426e-9cb9-9eaec90f5e2b-u1",
+ "output": {
+ "text": ["Your job results here"]
+ },
+ "status": "COMPLETED"
+}
+```
+
+Your webhook should return a `200` status code to acknowledge receipt. If the call fails, Runpod retries up to 2 more times with a 10-second delay between attempts.
### Execution policies
diff --git a/serverless/pricing.mdx b/serverless/pricing.mdx
index f91601086..40446c6bc 100644
--- a/serverless/pricing.mdx
+++ b/serverless/pricing.mdx
@@ -13,6 +13,8 @@ Runpod offers custom pricing plans for large scale and enterprise workloads. [Co
Serverless offers pay-per-second pricing with no upfront costs. You're billed from when a worker starts until it fully stops, rounded up to the nearest second.
+Per-GPU rates aren't listed in the documentation because they vary by GPU type and change over time. Find current rates on the [Runpod pricing page](https://www.runpod.io/pricing) or in the Runpod console when creating an endpoint. To estimate a job's cost, multiply the listed hourly rate by your expected worker runtime.
+
## Worker types
| | Flex workers | Active workers |
From b7574f2872fef859c6302ce656373ed155d75681 Mon Sep 17 00:00:00 2001
From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com>
Date: Mon, 31 Aug 2026 13:15:42 +0000
Subject: [PATCH 2/2] docs: style pass on assistant-gap updates (active voice,
second person)
---
get-started/agent-skills.mdx | 2 +-
pods/pricing.mdx | 2 +-
serverless/endpoints/endpoint-configurations.mdx | 4 ++--
serverless/pricing.mdx | 2 +-
4 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/get-started/agent-skills.mdx b/get-started/agent-skills.mdx
index 53131a239..87ac65637 100644
--- a/get-started/agent-skills.mdx
+++ b/get-started/agent-skills.mdx
@@ -10,7 +10,7 @@ The Runpod skills plugin teaches your coding agent how to run GPU workloads on R
It works with [Claude Code](https://code.claude.com/docs), [Codex](https://github.com/openai/codex), [Cursor](https://cursor.com/), [GitHub Copilot](https://github.com/features/copilot), [Windsurf](https://codeium.com/windsurf), [Cline](https://github.com/cline/cline), and [many other AI agents](https://www.skills.sh/).
-The skills plugin requires a coding agent with terminal access and Node.js installed (the install command uses `npx`). It can't be installed into chat apps like the ChatGPT or Claude mobile apps, which don't provide a shell.
+The skills plugin requires a coding agent with terminal access and Node.js installed (the install command uses `npx`). You can't install it in chat apps like the ChatGPT or Claude mobile apps, which don't provide a shell.
## Quick start
diff --git a/pods/pricing.mdx b/pods/pricing.mdx
index 21b924944..530495be1 100644
--- a/pods/pricing.mdx
+++ b/pods/pricing.mdx
@@ -15,7 +15,7 @@ Runpod offers custom pricing plans for large scale and enterprise workloads. [Co
Pods are billed by the second for compute and storage, with no fees for data ingress or egress.
-Per-GPU hourly rates and savings plan discounts aren't listed in the documentation because they vary by GPU type and cloud type (Secure or Community) and change over time. Find current rates on the [Runpod pricing page](https://www.runpod.io/pricing) or in the [Runpod console](https://www.console.runpod.io/pods) during Pod deployment.
+The documentation doesn't list per-GPU hourly rates or savings plan discounts because they vary by GPU type and cloud type (Secure or Community) and change over time. Find current rates on the [Runpod pricing page](https://www.runpod.io/pricing) or in the [Runpod console](https://www.console.runpod.io/pods) during Pod deployment.
## Pricing options
diff --git a/serverless/endpoints/endpoint-configurations.mdx b/serverless/endpoints/endpoint-configurations.mdx
index 957b50938..46514c2b3 100644
--- a/serverless/endpoints/endpoint-configurations.mdx
+++ b/serverless/endpoints/endpoint-configurations.mdx
@@ -119,9 +119,9 @@ Results are permanently deleted after retention expires.
### FlashBoot
Reduces cold starts by retaining worker state after spin-down, allowing faster "revival" than fresh boots. Most effective on endpoints with consistent traffic where workers frequently cycle between active and idle.
-Both new GPU and CPU endpoints will have FlashBoot enabled by default, and you can edit existing endpoints to enable or disable FlashBoot.
+New GPU and CPU endpoints have FlashBoot enabled by default, and you can edit existing endpoints to enable or disable it.
-State retention is opportunistic, not guaranteed. If a worker's retained state is no longer available, for example after a long period without traffic, the worker starts with a standard cold start instead. To guarantee low latency, set [active workers](#active-workers) above zero.
+State retention is best-effort, not guaranteed. If a worker's retained state is no longer available, for example after a long period without traffic, the worker starts with a standard cold start instead. To guarantee low latency, set [active workers](#active-workers) above zero.
When managing endpoints through the [REST API](/api-reference-v2/serverless/create-a-serverless-endpoint), the `flashboot` field accepts three modes: `OFF`, `FLASHBOOT`, and `PRIORITY_FLASHBOOT` (FlashBoot with priority capacity).
diff --git a/serverless/pricing.mdx b/serverless/pricing.mdx
index 40446c6bc..35ecb1548 100644
--- a/serverless/pricing.mdx
+++ b/serverless/pricing.mdx
@@ -13,7 +13,7 @@ Runpod offers custom pricing plans for large scale and enterprise workloads. [Co
Serverless offers pay-per-second pricing with no upfront costs. You're billed from when a worker starts until it fully stops, rounded up to the nearest second.
-Per-GPU rates aren't listed in the documentation because they vary by GPU type and change over time. Find current rates on the [Runpod pricing page](https://www.runpod.io/pricing) or in the Runpod console when creating an endpoint. To estimate a job's cost, multiply the listed hourly rate by your expected worker runtime.
+The documentation doesn't list per-GPU rates because they vary by GPU type and change over time. Find current rates on the [Runpod pricing page](https://www.runpod.io/pricing) or in the Runpod console when creating an endpoint. To estimate a job's cost, multiply the listed hourly rate by your expected worker runtime.
## Worker types