Yz

gemma

Gemma is a family of lightweight, state-of-the-art open models built by Google DeepMind. Updated to version 1.1

2b 7b
Hızlı Kurulum (Ollama kuruluysa) 
ollama run gemma

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 5.0GB 8K Text ollama run gemma:latest
instruct 5.0GB 8K Text ollama run gemma:instruct
text 5.2GB 8K Text ollama run gemma:text
v1.1 5.0GB 8K Text ollama run gemma:v1.1
2b 1.7GB 8K Text ollama run gemma:2b
7b 5.0GB 8K Text ollama run gemma:7b
2b-instruct 1.6GB 8K Text ollama run gemma:2b-instruct
2b-instruct-q2_K 1.3GB 8K Text ollama run gemma:2b-instruct-q2_K
2b-instruct-v1.1-q2_K 1.2GB 8K Text ollama run gemma:2b-instruct-v1.1-q2_K
2b-instruct-q3_K_S 1.4GB 8K Text ollama run gemma:2b-instruct-q3_K_S
2b-instruct-v1.1-q3_K_S 1.3GB 8K Text ollama run gemma:2b-instruct-v1.1-q3_K_S
2b-instruct-q3_K_M 1.5GB 8K Text ollama run gemma:2b-instruct-q3_K_M
2b-instruct-v1.1-q3_K_M 1.4GB 8K Text ollama run gemma:2b-instruct-v1.1-q3_K_M
2b-instruct-q3_K_L 1.6GB 8K Text ollama run gemma:2b-instruct-q3_K_L
2b-instruct-v1.1-q3_K_L 1.5GB 8K Text ollama run gemma:2b-instruct-v1.1-q3_K_L
2b-instruct-q4_0 1.7GB 8K Text ollama run gemma:2b-instruct-q4_0
2b-instruct-v1.1-q4_0 1.6GB 8K Text ollama run gemma:2b-instruct-v1.1-q4_0
2b-instruct-q4_1 1.8GB 8K Text ollama run gemma:2b-instruct-q4_1
2b-instruct-v1.1-q4_1 1.7GB 8K Text ollama run gemma:2b-instruct-v1.1-q4_1
2b-instruct-q4_K_S 1.7GB 8K Text ollama run gemma:2b-instruct-q4_K_S
2b-instruct-v1.1-q4_K_S 1.6GB 8K Text ollama run gemma:2b-instruct-v1.1-q4_K_S
2b-instruct-q4_K_M 1.8GB 8K Text ollama run gemma:2b-instruct-q4_K_M
2b-instruct-v1.1-q4_K_M 1.6GB 8K Text ollama run gemma:2b-instruct-v1.1-q4_K_M
2b-instruct-q5_0 1.9GB 8K Text ollama run gemma:2b-instruct-q5_0
2b-instruct-v1.1-q5_0 1.8GB 8K Text ollama run gemma:2b-instruct-v1.1-q5_0

77 varyant daha var — tümü için ollama.com.

Model Detayları ve Benchmarklar (kaynak: ollama.com)

Note: this model requires Ollama 0.1.26 or later. Download it here.

Gemma is a new open model developed by Google and its DeepMind team. It’s inspired by Gemini models at Google.

Gemma is available in both 2b and 7b parameter sizes:

  • ollama run gemma:2b
  • ollama run gemma:7b (default)

The models undergo training on a diverse dataset of web documents to expose them to a wide range of linguistic styles, topics, and vocabularies. This includes code to learn syntax and patterns of programming languages, as well as mathematical text to grasp logical reasoning.

To ensure the safety of the model, the team employed various data cleaning and filtering techniques, including rigorous filtering for CSAM (child sexual abuse material), sensitive data filtering, and filtering based on content quality in compliance with Google’s policies.

Reference

Google Gemma