Hugging Face Compatible

Drop-in compatibility with the Hugging Face ecosystem.

Transparent Proxy Mode

Switch to private hosting with zero code changes. MatrixHub provides a drop-in endpoint replacement that your existing tools already understand.

Terminal

# 1. Point your environment to MatrixHub
export HF_ENDPOINT="http://matrixhub.internal.company.com:3001"

# 2. Run your existing inference engine (e.g. vLLM)
python -m vllm.entrypoints.openai.api_server \
  --model deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct

# MatrixHub automatically proxies the request, caches the model locally,
# and serves it to all subsequent nodes at 10Gbps+ speeds.

🔌

API Compatibility Boundary

We meticulously implement the subset of the Hugging Face API required by vLLM, SGLang, and common Hugging Face clients. No bloated feature-parity, just the infrastructure layer enterprise inference teams actually need.

🗂️

Native Artifact Support

Seamlessly interact with PyTorch bins, Safetensors, GGML/GGUF, and tokenizers. MatrixHub understands model directory structures natively, allowing for optimized chunking and delivery.