Skip to main content

My Models

Overview​

My Models page is a centralized registry for managing your model assets. On this page, you can view all model assets, upload local model weights, and import models from Hugging Face.

Model list​

The Model List shows all your model assets. From this page, you can upload new models, view detailed model configurations, and quickly deploy models to active inference endpoints.


model-list


Upload model​

Click Upload Model to register a new model. Fill in these key fields:

  • Name: Model name.
  • Type (optional): Select the model type from the dropdown(LLM, VLM, Embedding, Reranker, Classification, Reward, ASR).
  • Parameter Size (optional): Model parameter size (e.g., 7B, 13B, 70B).
  • Architecture (optional): Model architecture (e.g., Qwen2ForCausalLM).
  • Description (optional): Optional notes about the model. Maximum 1024 characters.
  • Model folder: Select a Hugging Face model directory from your local machine.

Serving configuration​

Optional SGLang runtime defaults stored with this model.

  • Configuration JSON (optional): JSON configuration for the SGLang runtime. Example:
{
  "schema_version": 1,
  "dtype": "auto",
  "trust_remote_code": false
}

upload


Next steps​

Deploy Your Model

Deploy a model from the gallery or upload your own model weights.

Learn More
Fine-Tune Your Model

Adapt a base model to your domain using SFT, DPO, or CPT.

Learn More