---
title: "Nvidia Nim Proxy Route"
method: POST
path: "/nvidia_nim/{endpoint}"
tags: ["NVIDIA NIM Pass-through", "pass-through"]
---

# Nvidia Nim Proxy Route

`POST /nvidia_nim/{endpoint}`

Relay a native NVIDIA NIM request through a LiteLLM model group.

`{PROXY_BASE_URL}/nvidia_nim/{model_group}/v1/infer` forwards the body unchanged to the deployment's
`api_base`, so object detection and OCR NIMs whose payload carries no `model` field still go through
virtual key auth, model access checks, and spend logging.

## Path parameters

- `endpoint` string, required

## Response `200`

Successful Response

- unknown

## Other responses

- `422` — Validation Error

## Changes

- **2026-10-05** `611a0ed75fe8` — 3 info
  - api tag `NVIDIA NIM Pass-through` added
  - api tag `pass-through` added
  - api tag `llm_passthrough` removed
- **2026-10-04** `51c2ca10dc0b` — 3 info
  - api tag `llm_passthrough` added
  - api tag `NVIDIA NIM Pass-through` removed
  - api tag `pass-through` removed
- **2026-10-03** `c94c9b11f95e` — 3 info
  - api tag `NVIDIA NIM Pass-through` added
  - api tag `pass-through` added
  - api tag `llm_passthrough` removed
- **2026-09-30** `85ac4287140e` — 1 info
  - endpoint added

[Change history](https://skmtc.dev/flock/apis/litellm-api/changes/nvidia_nim/:endpoint/post.md)

---

[API](https://skmtc.dev/flock/apis/litellm-api.md) · [All operations](https://skmtc.dev/flock/apis/litellm-api/llms.txt) · [OpenAPI document](https://skmtc.dev/flock/apis/litellm-api/revisions/ee3a9ebe33fd?raw)