Yz

granite-embedding

The IBM Granite Embedding 30M and 278M models models are text-only dense biencoder embedding models, with 30M available in English only and 278M serving multilingual use cases.

Embedding 30m 278m
Hızlı Kurulum (Ollama kuruluysa) 
ollama run granite-embedding

Ollama kurulu değil mi? ollama.com/download — Windows, macOS ve Linux için ücretsiz. İlk çalıştırmada model indirilir, sonrası tamamen çevrimdışıdır.

Varyantlar

Boyut büyüdükçe kalite artar, donanım ihtiyacı yükselir. Başlangıç için küçük varyantı deneyin.

EtiketBoyutBağlamGirdiKomut
latest 63MB 512 Text ollama run granite-embedding:latest
30m 63MB 512 Text ollama run granite-embedding:30m
278m 563MB 512 Text ollama run granite-embedding:278m
30m-en 63MB 512 Text ollama run granite-embedding:30m-en
30m-en-fp16 63MB 512 Text ollama run granite-embedding:30m-en-fp16
278m-fp16 563MB 512 Text ollama run granite-embedding:278m-fp16

Model Detayları ve Benchmarklar (kaynak: ollama.com)

Granite embedding models

The IBM Granite Embedding 30M and 278M models are text-only dense biencoder embedding models, with 30M available in English only and 278M serving multilingual use cases. These models are designed to produce fixed length vector representations for a given text chunk, which can be used for text similarity, retrieval, and search applications.

Parameter Sizes

30M:

ollama run granite-embedding:30m

278M:

ollama run granite-embedding:278m

Supported Languages

30M: English 278M: English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, Chinese (Simplified)

Granite dense models

The Granite dense models are available in 2B and 8B parameter sizes designed to support tool-based use cases and for retrieval augmented generation (RAG), streamlining code generation, translation and bug fixing.

See model page

Granite mixture of experts models

The Granite mixture of experts models are available in 1B and 3B parameter sizes designed for low latency usage.

See model page

Learn more