ai-on-gke
gemma_pytorch
| ai-on-gke | gemma_pytorch | |
|---|---|---|
| 2 | 6 | |
| 168 | 5,084 | |
| 16.1% | 1.2% | |
| 9.8 | 7.8 | |
| 2 days ago | 4 days ago | |
| Jupyter Notebook | Python | |
| Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
ai-on-gke
-
Gemma: New Open Models
There is a lot of work to make the actual infrastructure and lower level management of lots and lots of GPUs/TPUs open as well - my team focuses on making the infrastructure bit at least a bit more approachable on GKE and Kubernetes.
https://github.com/GoogleCloudPlatform/ai-on-gke/tree/main
and
https://github.com/google/xpk (a bit more focused on HPC, but includes AI)
and
https://github.com/stas00/ml-engineering (not associated with GKE, but describes training with SLURM)
The actual training is still a bit of a small pool of very experienced people, but it's getting better. And every day serving models gets that much faster - you can often simply draft on Triton and TensorRT-LLM or vLLM and see significant wins month to month.
- Reference Architecture for ML Training and Batch on GKE with Kueue
gemma_pytorch
-
Getting Started with Gemma Models
Gemma is a family of lightweight, open-source machine learning models developed by Google AI. These models are designed to be accessible and efficient, making AI development more available for a broad range of users. Released on February 21st, 2024, Gemma is built from the same research and technology that was used to create the Gemini models. Amongst the key features, which are being lightweight and open-source, Gemma is also text-based. It excels in tasks like text summarization, question answering, and reasoning.
-
Gemma doesn't suck anymore – 8 bug fixes
Here are the missing links:
* Gemma, a family of open models from Google: https://ai.google.dev/gemma
* Unsloth is a tool/method for training models faster (IIUC): https://github.com/unslothai/unsloth
- The official PyTorch implementation of Google's Gemma models
-
Gemma: New Open Models
The release page has comparisons to Mistral everywhere: https://ai.google.dev/gemma
What are some alternatives?
llama.cpp - LLM inference in C/C++
gemma - Open weights LLM from Google DeepMind.
ollama - Get up and running with Llama 3, Mistral, Gemma, and other large language models.
gemma.cpp - lightweight, standalone C++ inference engine for Google's Gemma models.
text-to-text-transfer-transformer - Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"