docs: add Neon AI Gateway integration guide

docs/docs.json

@@ -192,6 +192,7 @@
                   "getting-started/integration-method/hyperbolic",
                   "getting-started/integration-method/mistral",
                   "getting-started/integration-method/nebius",
+                  "getting-started/integration-method/neon",
                   "getting-started/integration-method/novita",
                   {
                     "group": "Nvidia",

docs/getting-started/integration-method/neon.mdx

@@ -0,0 +1,130 @@
+---
+title: "Neon AI Gateway Integration"
+sidebarTitle: "Neon AI Gateway"
+description: "Log Neon AI Gateway requests in Helicone by routing them through the Helicone Gateway. Neon serves an OpenAI-compatible endpoint, so the standard proxy headers apply."
+"twitter:title": "Neon AI Gateway Integration - Helicone OSS LLM Observability"
+---
+
+[Neon AI Gateway](https://neon.com/docs/ai-gateway/overview) is an OpenAI-compatible inference endpoint provided by Neon. A single Neon credential gives access to models from OpenAI, Google, Meta, Databricks, and Alibaba.
+
+<Note>
+  Neon AI Gateway is in beta. It requires a paid Neon plan and a project in the AWS US East (Ohio) region (`aws-us-east-2`).
+</Note>
+
+Neon is not one of Helicone's registered providers, so it is not selectable in the [AI Gateway](/gateway/overview) model registry. Route it through the [Helicone Gateway](/getting-started/integration-method/gateway) instead, which proxies any OpenAI-compatible host you name in a header.
+
+## Gateway integration
+
+<Steps>
+  <Step title="Create a Helicone account">
+    Log into [Helicone](https://www.helicone.ai) or create an account, then generate an [API key](https://helicone.ai/developer).
+  </Step>
+  <Step title="Create a Neon credential">
+    In the Neon Console, select your branch, open **Credentials** under **APP BACKEND**, and create a credential with the `ai_gateway:invoke` scope. See [AI Gateway authentication](https://neon.com/docs/ai-gateway/authentication).
+  </Step>
+  <Step title="Find your branch host">
+    Each Neon branch has its own gateway host, shown in the Neon Console as `NEON_AI_GATEWAY_BASE_URL` and written below as `https://<your-neon-branch-host>`. There is no shared Neon hostname to point at.
+  </Step>
+  <Step title="Set the environment variables">
+```bash
+HELICONE_API_KEY="<your-helicone-api-key>"
+NEON_AI_GATEWAY_TOKEN="nt_live_..."
+NEON_AI_GATEWAY_BASE_URL="https://<your-neon-branch-host>"
+```
+  </Step>
+  <Step title="Point requests at the Helicone Gateway">
+
+Replace the Neon branch host with `https://gateway.helicone.ai` and pass the branch host in `Helicone-Target-Url`:
+
+`$NEON_AI_GATEWAY_BASE_URL/v1/chat/completions` -> `https://gateway.helicone.ai/v1/chat/completions`
+
+```
+Helicone-Auth: `Bearer ${HELICONE_API_KEY}`
+Helicone-Target-Url: `${NEON_AI_GATEWAY_BASE_URL}`
+Authorization: `Bearer ${NEON_AI_GATEWAY_TOKEN}`
+```
+
+  </Step>
+</Steps>
+
+## Example
+
+<Tabs>
+  <Tab title="cURL">
+
+```bash
+curl --request POST \
+  --url https://gateway.helicone.ai/v1/chat/completions \
+  --header "Authorization: Bearer $NEON_AI_GATEWAY_TOKEN" \
+  --header "Helicone-Auth: Bearer $HELICONE_API_KEY" \
+  --header "Helicone-Target-Url: $NEON_AI_GATEWAY_BASE_URL" \
+  --header "Content-Type: application/json" \
+  --data '{
+    "model": "gpt-5-mini",
+    "messages": [{ "role": "user", "content": "What is the capital of France?" }]
+  }'
+```
+
+  </Tab>
+  <Tab title="Node.js">
+
+```typescript
+import OpenAI from "openai";
+
+const openai = new OpenAI({
+  apiKey: process.env.NEON_AI_GATEWAY_TOKEN,
+  baseURL: "https://gateway.helicone.ai/v1",
+  defaultHeaders: {
+    "Helicone-Auth": `Bearer ${process.env.HELICONE_API_KEY}`,
+    "Helicone-Target-Url": process.env.NEON_AI_GATEWAY_BASE_URL,
+  },
+});
+
+const completion = await openai.chat.completions.create({
+  model: "gpt-5-mini",
+  messages: [{ role: "user", content: "What is the capital of France?" }],
+});
+```
+
+  </Tab>
+  <Tab title="Python">
+
+```python
+import os
+from openai import OpenAI
+
+client = OpenAI(
+    api_key=os.environ["NEON_AI_GATEWAY_TOKEN"],
+    base_url="https://gateway.helicone.ai/v1",
+    default_headers={
+        "Helicone-Auth": f"Bearer {os.environ['HELICONE_API_KEY']}",
+        "Helicone-Target-Url": os.environ["NEON_AI_GATEWAY_BASE_URL"],
+    },
+)
+
+completion = client.chat.completions.create(
+    model="gpt-5-mini",
+    messages=[{"role": "user", "content": "What is the capital of France?"}],
+)
+```
+
+  </Tab>
+</Tabs>
+
+Streaming works the same way: add `stream: true` and Neon forwards `text/event-stream` responses through the gateway.
+
+## Rate limits on Neon branch hosts
+
+Neon branch hosts are not on Helicone's [approved domain list](/getting-started/integration-method/gateway#approved-domains), and each Neon branch has a different hostname, so requests fall under the unapproved-domain limits of 10,000 requests per day and 1 request per second. To lift those limits for your branch hosts, contact engineering@helicone.ai or ask on [Discord](https://discord.gg/zsSTcH2qhG).
+
+If you need higher throughput before a domain is approved, use a [manual logger](/getting-started/integration-method/manual-logger-typescript) instead. It logs requests asynchronously and leaves the call path to Neon untouched.
+
+## Models
+
+Neon uses short model IDs such as `gpt-5-mini`, `gemini-3-flash`, and `qwen3-next-80b-a3b-instruct`. List what your branch can reach with `GET $NEON_AI_GATEWAY_BASE_URL/v1/models`, or browse the [Neon model catalog](https://neon.com/docs/ai-gateway/models). Model IDs, pricing, and capabilities are also published as the [`neon` provider on Models.dev](https://models.dev/providers/neon/).
+
+<Info>
+  Cost calculation is not enabled for Neon branch hosts, so requests logged this way show token counts without a cost figure. Helicone's automated mapper still resolves the request and response schema, which can take up to 24 hours to appear in the UI.
+</Info>
+
+For more on the headers used above, see [Helicone Headers](https://docs.helicone.ai/helicone-headers/header-directory#utilizing-headers).