Ross ROSS = Recommend OSS · open-source software intelligence for agents

huggingface/text-embeddings-inference

A blazing fast inference solution for text embeddings models observed · 2026-08-28

github.com/huggingface/text-embeddings-inference · homepage · Rust · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

84/100

  • Activity 94
  • Release rhythm 76
  • Longevity 75
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: 26
  • age_days: 1055
  • days_rel: 163
  • days_push: 40
  • n_releases_24m: 16

Full methodology

Adoption not part of the score

5028 stars · 426 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Text Embeddings Inference (TEI) is a Rust-based toolkit from Hugging Face for deploying and serving open-source text embedding, re-ranker, and sequence classification models at high performance. It supports dynamic batching, Flash Attention, Safetensors/ONNX weight loading, and production features like Prometheus metrics and OpenTelemetry tracing.

Use cases

  • serve embedding models for a RAG pipeline
  • host a fast embeddings API endpoint for semantic search
  • deploy re-ranker models behind an HTTP or gRPC API
  • run embeddings inference on GPU, CPU, or Apple Silicon locally
  • self-host embeddings instead of paying per-token API costs
  • batch-embed large document collections efficiently

When to choose

  • you need production-grade, low-latency serving of supported embedding models like BGE, GTE, E5, Qwen3, or Nomic
  • you want small Docker images, fast boot times, and serverless-friendly deployment
  • you need dynamic batching, tracing, and metrics out of the box
  • you want to serve private or gated Hugging Face models on your own infrastructure

When to avoid

  • you need to serve embedding model architectures not on the supported list
  • you only need occasional embeddings and prefer a hosted API without infrastructure
  • you need text generation inference rather than embeddings (use TGI instead)
  • you want a Python-native stack for experimentation rather than a Rust server

Facets

service · maturity active

llm-inference http-server machine-learning rag machine-learning large-language-models artificial-intelligence developer-tools rust self-hosted embeddings huggingface rerankers sequence-classification flash-attention safetensors onnx grpc prometheus open-telemetry retrieval-augmented-generation docker linux macos gpu web-server

10 sources

Member repositories

RepositoryRoleHealth v2
huggingface/text-embeddings-inferencemain84

For agents

markdown · JSON · MCP: product_card(name="huggingface/text-embeddings-inference")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem