Google · Ollama

functiongemma

FunctionGemma is a specialized version of Google's Gemma 3 270M model fine-tuned explicitly for function calling.

tools 270m

126.6K Pulls 4 Tags Updated 3 months ago

embeddinggemma

EmbeddingGemma is a 300M parameter embedding model from Google.

embedding 300m

884K Pulls 5 Tags Updated 6 months ago

gemma2

Google Gemma 2 is a high-performing and efficient model available in three sizes: 2B, 9B, and 27B.

2b 9b 27b

19.6M Pulls 94 Tags Updated 1 year ago

gemma

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

2b 7b

6.5M Pulls 102 Tags Updated 1 year ago

glm-ocr

GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.

vision tools

175K Pulls 3 Tags Updated 1 month ago

rinex20/translategemma3

An optimized version of Google's TranslateGemma-12B-it (Gemma 3) designed for high-fidelity translation. This build features hard-coded Temperature=0.1 and English Anchor support to eliminate output redundancy and maximize accuracy.

12b

32.9K Pulls 1 Tag Updated 2 months ago

dcarrascosa/medgemma-1.5-4b-it

This repo packages Google’s google/medgemma-1.5-4b-it (MedGemma 1.5, 4B, instruction-tuned, multimodal) for Ollama, exposing vision + text inference and common Ollama features (structured JSON output, tool calling via your app, etc.)

vision

755 Pulls 3 Tags Updated 1 month ago

MedAIBase/TranslateGemma

TranslateGemma is a family of lightweight, state-of-the-art open translation models from Google, based on the Gemma 3 family of models. TranslateGemma models are designed to handle translation tasks across 55 languages.

vision 4b 12b 27b

514 Pulls 12 Tags Updated 2 months ago

aliafshar/gemma3-it-qat-tools

Gemma3 instruction-tuned and quantized with tool/functions using a faithful representation of the Google DeepMind recommended tool calling methodology. Runs remarkably well on old GPUs and CPU.

vision tools 1b 4b 12b 27b

12.6K Pulls 5 Tags Updated 9 months ago

urtu/Spencer-AI-Butler

Ever wanted a butler, but you were too poor to pay for one? Well, with this model, you can make that dream come true, even with as little as 8 GB of RAM. Based on Google's open-sourced model Gemma (with 8 billion parameters), it is a conversational LLM.

vision

50 Pulls 1 Tag Updated 3 months ago

itzune/kimu

Kimu is an instruction large language model (LLM) tailored specifically for the Basque language built from Google's Gemma-2 foundational and Gemma-2 instruct models.

2b 9b

35 Pulls 5 Tags Updated 5 months ago

bjoernb/gemma3n-e4b

Google Gemma 3n | Edge AI with tool support. Designed for consumer devices: efficient, local, tool-enabled

tools

320 Pulls 1 Tag Updated 8 months ago

jwang580/medgemma_27b_text_it

MedGemma is a Gemma 3 variants that are trained for performance on medical text and image comprehension developed by Google. 27B is text only. For all MedGemma 27B results, test-time scaling is used to improve performance. This is the full version.

337 Pulls 1 Tag Updated 9 months ago

antony66/gemma3-tools

Enhanced function (tool) calling capabilities with modified template as supposed by Philipp Schmid (Google DeepMind) for gemma3 models.

vision tools 27b

374 Pulls 1 Tag Updated 11 months ago

SimonPu/gemma3

The Google Gemma 3 models are multimodal—processing text and images—and feature a 128K context window with support for over 140 languages....

vision

264 Pulls 4 Tags Updated 11 months ago

sam860/gemma3n

Google's MatFormer addition to the gemma3 family

e2b

186 Pulls 6 Tags Updated 6 months ago

second_constantine/gemma3-qat

Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. (quantized IQ4_XS)

vision 27b

149 Pulls 1 Tag Updated 6 months ago

bjoernb/gemma3n-e2b

Google Gemma 3n | Edge AI with tool support Designed for consumer devices: efficient, local, tool-enabled

tools

134 Pulls 1 Tag Updated 8 months ago

second_constantine/gemma3

Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. (quantized UD-Q4_K_XL)

vision 27b

117 Pulls 1 Tag Updated 6 months ago

Tharusha_Dilhara_Jayadeera/singemma

A Sinhala-adapted version of Google’s Gemma 3 4B, continually pre-trained on 10.7M Sinhala sentences with a custom 16k vocabulary using 4-bit LoRA.

110 Pulls 1 Tag Updated 6 months ago