search
Gemma
Trends
- 1Google releases EmbeddingGemma 2, a lightweight open multimodal embedding modelโEmbeddingGemma 2: An open, lightweight multimodal embedding model
Google has announced EmbeddingGemma 2, an open, lightweight multimodal embedding model aimed at developers building search, retrieval and AI applications. The model is designed to run efficiently on modest hardware while handling text and other modalities. Discussion among developers is focused on how it compares with rival open embedding models and what its small size means for on-device and low-cost deployments.
- 2
Google has released EmbeddingGemma 2, a new version of its open embedding model for developers. The announcement, published on Google's blog, drew attention and discussion among the developer community, with commenters trading notes on the model's capabilities and practical uses in AI applications.
- 3
Google has released EmbeddingGemma 2, a new embedding model the company says can run directly on devices and handle multimodal inputs. The launch is being discussed among developers and AI watchers as part of the push to bring capable AI models to phones and other consumer hardware without relying on cloud servers. Reaction online has focused on what the model means for privacy, latency and on-device applications.
- 4Hacktoberfest AI Challenge opens with 'Touch Grass' themeโThis is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass. ... # devchallenge # hf26challe
The first week of the Hacktoberfest Open-Source AI Challenge is underway, with the theme 'Touch Grass' prompting projects that push people outdoors. Entries include Touch Grass Birder, a bird call identifier that works offline; Outside Window, a local AI agent built on Gemma designed to get users off their screens; and OutingFold, a pocket field guide. Builders are sharing submissions under the #hf26challenge hashtag.
- 5Run LLMs Locally adds EmbeddingGemma2 multimodal embeddingsโNew update: Run LLMs Locally Added EmbeddingGemma2 using llama.cpp. It includes multi modality, allowing to generate emb
A developer released an update to Run LLMs Locally adding EmbeddingGemma2 support via llama.cpp. The new version handles multimodal inputs, generating embedding vectors from text, images, video, and audio, which makes it possible to build local search engines over documents and photos. The project is available on GitHub, and it is drawing attention among people interested in running AI tools offline on their own hardware.
- 6AeroGemma app tells you if it's a good day to go outsideโAeroGemma: Checking If You Can Actually Touch Grass Today ๐ฟ I spend most of my day in front of a... # devchallenge # hf2
A developer has built AeroGemma, an open-source tool that combines AI with local air quality data to tell users whether they should go outside and 'touch grass' today. Built around Google's Gemma model, the project was created for a hackathon challenge and shared with the developer community, drawing attention for its playful take on a common internet phrase applied to real environmental data.
- 7
Google has announced EmbeddingGemma 2, the second generation of its compact open embedding model for text. The release is drawing attention among developers, who are discussing its performance and how it compares with other embedding models for search, retrieval and AI applications. Details on benchmarks and availability are circulating as the community assesses the update.
- 8Hacktoberfest AI Challenge opens with Touch Grass themeโThis is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass ... # devchallenge # hf26challen
A new submission has been entered for Week 1 of the Hacktoberfest Open-Source AI Challenge, themed Touch Grass. The project, called Paper Trail, uses Google's Gemma AI model to generate field cards documenting a user's walk. The challenge invites developers to build open-source AI tools, with entries being shared across developer communities as the competition's first week gets underway.
- 9Developer builds offline AI planner that picks your best hour outdoorsโSay what you want to do outside. Gemma, on your own computer, reads it, code picks the hour from an open forecast, and y
A developer has built a small open-source tool that lets users type what they want to do outside; Google's Gemma model runs locally on the user's own computer, code picks the best hour from an open weather forecast, and the result comes back as a reminder and a pocket card that works without any signal. The project is being shared as part of an open-source development challenge.
Repos
- FeSens/openTPU An open-source AI accelerator, developed by AI: RTL, ISA, simulator, compiler and profiler in one repo. Runs Qwen3, LFM2
- firelex/jeff Millisecond decisions, any domain: a 0.8B open "System 1" model that picks between your options with calibrate