Technology

granite-embedding 278m

IBM's high-efficiency 278M parameter model optimized for multilingual text embeddings and high-performance RAG workflows.

The granite-embedding 278m model delivers a lean, high-performance solution for dense vector retrieval across 38 languages. Built on a transformer architecture with a 512-token context window, it balances a compact 278-million parameter footprint with state-of-the-art accuracy on the MTEB benchmark. IBM engineered this model specifically for Retrieval-Augmented Generation (RAG) tasks, ensuring low-latency inference and precise semantic search capabilities for enterprise AI pipelines.

https://huggingface.co/ibm-granite/granite-embedding-278m-multilingual

What builders pair with granite-embedding 278m

Projects using both technologies. Select a pairing to see a project.

Built with granite-embedding 278m

AI agents 1 shared project dice rolls 1 shared project Docker 1 shared project Jan-nano-gguf 1 shared project monsters 1 shared project Qwen-2 1 shared project

2 more pairings

RAG 1 shared project Scratch 1 shared project

Pairing: AI agents

Photo from the event — Event photo

Compose and Dragons: Tiny Language Models in Action

Paris · April 21, 2026

view project | view AI agents

Recent Talks & Demos

Showing 1-1 of 1

Members-Only

Sign in to see who built these projects

Sign in View FAQ

Compose and Dragons: Tiny Language Models in Action

Docker Jan-nano-gguf