Admin Management

Create Docker Image From Huggingface

Get runtime configuration for deploying a HuggingFace model with vLLM.

This endpoint:

  1. Analyzes the HuggingFace model to extract specs
  2. Determines optimal GPU configurations (single GPU vs multi-GPU)
  3. Returns the pre-built vLLM base image with environment variables
  4. NO Docker build required - uses runtime configuration!

Args: request: CreateDockerImageRequest with HuggingFace URL

Returns: Model specifications, GPU recommendations, and runtime configuration

post/v1/admin/create-docker-image-from-huggingface

Request

  • Base URL: https://api.oneinfer.ai (the document declares no server; this is the origin it was published from)
  • URL: https://api.oneinfer.ai/v1/admin/create-docker-image-from-huggingface
  • Auth: HTTP bearer

Request body

huggingface_model_urlstring required

HuggingFace model URL (e.g., https://huggingface.co/openai/gpt-oss-20b)

dockerhub_usernamestring required

Docker Hub username

dockerhub_passwordstring required

Docker Hub password/token

image_namestring nullable

Custom image name (auto-generated if not provided)

tagstring

Docker image tag

build_onlyboolean

Only build, don't push to Docker Hub

hf_tokenstring nullable

HuggingFace token for gated models (e.g., Llama, Mistral). Required for models that need license acceptance.

Response

Successful Response

job_idstring required

Unique job ID for tracking build progress

statusstring required

Status: queued/building/pushing/success/failed/docker_not_available

image_namestring required

Full Docker image name

build_logstring nullable

Build log output (partial for in-progress jobs)

docker_pull_commandstring nullable

Command to pull the image

estimated_build_time_minutesinteger required

Estimated build time in minutes

progress_percentageinteger nullable

Build progress percentage (0-100)

current_stepstring nullable

Current build step description

Changes

No recorded changes to this operation across all 1 revision of this API.