Welcome to the Gemmaverse
Explore Gemma models crafted by the community
Case studies
-
SCBX Group improves the performance, size, cost-efficiency, and Thai language proficiency of its Typhoon model with Gemma 3 12B.
-
Gemma 3 4B helps Levro build a model training platform with lower inference costs.
-
Crane AI Labs bridges regional language and connectivity gaps with Swahili Gemma 1B.
-
The Ministry of Economy, Ecology and Agriculture of Ukraine digitizes its licensing process using Gemma.
-
Adaptive ML aids SK Telecom in creating a version of Gemma that can moderate customer support at a fraction of the size, latency, and cost.
-
With the help of Gemma 2 and Gemma 3, Quarks delivers safer and more engaging experiences on their dating and connection apps.
-
Based on Gemma 3, Sarvam AI's model translates longform and structured content across India's diverse range of languages.
-
Brazilian organizations and startups collaborate to build GAIA, improving upon Gemma 3 performance in Brazilian Portuguese.
-
Roboflow aims to improve computer vision for consumer and enterprise use with its AI workflows, fully integrated with PaliGemma.
-
A powerful Japanese-focused LLM with Gemma 2.
-
Nexa AI uses Gemma as a foundation for OmniAudio, an audio-language model designed to maximize performance for edge applications.
-
AI Singapore releases the SEA-LION family of models, trained in 11 Southeast Asian languages to accurately reflect the region’s nuances, contexts, and cultural diversity.
-
INSAIT chooses Gemma for BgGPT, a Bulgarian-first model addressing developer needs and powering the public chat platform, BgGPT.ai.
alpha-ai/Medical-Diagnosis-COT-Gemma3-270M
A fine-tuned Gemma-3 270M model for medical question answering with explicit chain-of-thought (CoT)
CraneAILabs/swahili-gemma-1b
A fine-tuned Gemma 3 1B model specialized for English-to-Swahili translation and Swahili conversational AI
unsloth/gemma-3-270m-it-unsloth-bnb-4bit
A Gemma 3 270M model with 4-bit quantization for fast fine-tuning with Unsloth
AIDC-AI/Ovis1.6-Gemma2-9B
A 10.2B parameter multimodal large language model for image and text processing and generation
junnei/gemma-3-4b-it-speech
Open models extending the multimodal capabilities of Gemma 3 with speech processing
Menlo/voxel-representation-gemma3-4b
A Vision-Language Model (VLM) that processes 2D slices to efficiently extract semantic information from 3D voxel data
Join our Gemma community
Connect, explore, and share your knowledge with others in the ML model community