docs: add Auxen integration

[ai-auxen]

Adds documentation for routing Auxen private-model inference requests through Helicone's gateway for observability.

About Auxen: Auxen provisions per-instance OpenAI-compatible LLM endpoints on dedicated GPUs (Llama, Qwen, Mistral, Gemma, Phi-3, Mixtral). Each instance has its own base URL + API key.

Integration shape: Because Auxen instances are per-customer (no single global API host), this integration uses Helicone's generic gateway (gateway.helicone.ai) with the Helicone-Target-URL header pointing at the user's Auxen instance. No Helicone-side changes needed — purely docs.

Files:

• docs/integrations/auxen/javascript.mdx — Quickstart with OpenAI SDK + Vercel AI SDK provider examples
• docs/integrations/auxen/python.mdx — Quickstart with OpenAI Python SDK + Auxen Python SDK examples (async + streaming)
• docs/integrations/auxen/curl.mdx — Raw cURL recipe

Navigation: docs/docs.json updated — new Auxen group inserted alphabetically between Anyscale and AWS Bedrock.

Example usage (excerpt):

const auxen = new OpenAI({
  apiKey: process.env.AUXEN_API_KEY,
  baseURL: "https://gateway.helicone.ai/v1",
  defaultHeaders: {
    "Helicone-Auth": `Bearer ${process.env.HELICONE_API_KEY}`,
    "Helicone-Target-URL": process.env.AUXEN_BASE_URL,
  },
});

Links:

• Auxen homepage
• Auxen Python SDK (github.com/auxen-ai/auxen-python)
• Auxen MCP server (github.com/auxen-ai/mcp-server)

[greptile-apps]

Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.

[vercel]

@Samp2Alex is attempting to deploy a commit to the Helicone Team on Vercel.

A member of the Team first needs to authorize it.