Today's LLM & Language Models: Fastest-Growing Projects — May 28, 2026
Today's the LLM & Language Models space, there's a noticeable trend towards developing more versatile and accessible AI solutions that cater to diverse use cases, from enhancing user interactions with vision-based capabilities to optimizing large-scale language model training. One standout project is "minimind-o," which aims to create an all-encompassing 0.1B Omni model capable of various modalities like listening, speaking, and seeing.
jingyaogong/minimind-o (Growth Score: 42.50; Stars: 1,608) is a project focused on training a compact yet versatile multimodal language model from scratch, designed to handle auditory, visual, and linguistic tasks seamlessly. The high growth score and star count suggest strong community interest in the development of efficient, multitalented AI models that can perform across different modalities without requiring extensive computational resources.
JSingletonAI/dejavu (Growth Score: 33.85; Stars: 68) offers a memory solution for AI tools that allows users to maintain consistent data access and retrieval capabilities without the need for cloud storage or account management. The project's significant growth score indicates its appeal among developers seeking seamless integration of persistent, device-agnostic memory functionalities within various applications.
Continuum-AI-Corp/OrcaRouter-Lite (Growth Score: 24.46; Stars: 683) provides a self-hosted LLM router designed to be compatible with OpenAI services while supporting Bring Your Own Key and single-workspace configurations. The project's growing popularity, as indicated by its star count and commits, highlights the demand for robust yet customizable solutions that enhance security and control in AI environments.
wanshuiyin/ARIS-in-AI-Offer (Growth Score: 21.39; Stars: 121) offers bilingual interview cheat sheets tailored specifically for AI job seekers, auto-generated to be accessible across devices. The project's steady growth suggests its usefulness in preparing candidates for the technical aspects of AI interviews with comprehensive and conveniently formatted resources.
gonemedia/aipointer (Growth Score: 20.03; Stars: 232) is an innovative tool that allows users to interact directly with their cursor context through voice or text queries, leveraging multiple AI providers for real-time responses on various operating systems. The rising star count and growth score reflect the community's interest in integrating advanced language models into everyday computing tasks.
rahilp/second-brain-cloudflare (Growth Score: 18.61; Stars: 89) provides a self-hosted memory layer that can be accessed across different AI tools, enabling users to store and recall information effortlessly on Cloudflare's free tier. The growth in stars and commits underscores the project’s appeal as a scalable solution for managing persistent data without relying on proprietary cloud services.
chiennv2000/orthrus (Growth Score: 17.20; Stars: 372) focuses on accelerating lossless LLM inference through dual-view diffusion decoding, aiming to improve the efficiency of large model deployments. The high star count and growth score suggest that developers are interested in optimizing AI performance for real-world applications where computational resources are limited.
okturro/asr-rescore-bench (Growth Score: 16.00; Stars: 154) offers a benchmarking framework to evaluate LLM-based ASR n-best rescoring strategies, including various linguistic models and prompt techniques. The steady growth in stars indicates the project's importance for researchers and developers looking to refine automatic speech recognition systems.
sitodowubb/spatial-vqa-bench (Growth Score: 15.20; Stars: 146) introduces a benchmark specifically designed to assess spatial visual reasoning capabilities of multimodal LLMs, focusing on their ability to interpret and respond accurately to queries involving spatial relationships in images or scenes. The growth score reflects the growing interest in evaluating AI's proficiency in understanding complex visual contexts.
ahammadmejbah/Awesome-Datasets-Hub (Growth Score: 14.54; Stars: 132) compiles a comprehensive list of datasets relevant to LLMs, covering areas such as medical AI, NLP, multimodal learning, and more. The increasing number of stars highlights the resource's value for researchers and developers seeking curated datasets for training and evaluating advanced language models.
Today's report showcases projects that range from enhancing user interactions with advanced AI capabilities to optimizing model performance and providing essential resources for research and development in the field of large language models.
jingyaogong/minimind-o (Growth Score: 42.50; Stars: 1,608) is a project focused on training a compact yet versatile multimodal language model from scratch, designed to handle auditory, visual, and linguistic tasks seamlessly. The high growth score and star count suggest strong community interest in the development of efficient, multitalented AI models that can perform across different modalities without requiring extensive computational resources.
JSingletonAI/dejavu (Growth Score: 33.85; Stars: 68) offers a memory solution for AI tools that allows users to maintain consistent data access and retrieval capabilities without the need for cloud storage or account management. The project's significant growth score indicates its appeal among developers seeking seamless integration of persistent, device-agnostic memory functionalities within various applications.
Continuum-AI-Corp/OrcaRouter-Lite (Growth Score: 24.46; Stars: 683) provides a self-hosted LLM router designed to be compatible with OpenAI services while supporting Bring Your Own Key and single-workspace configurations. The project's growing popularity, as indicated by its star count and commits, highlights the demand for robust yet customizable solutions that enhance security and control in AI environments.
wanshuiyin/ARIS-in-AI-Offer (Growth Score: 21.39; Stars: 121) offers bilingual interview cheat sheets tailored specifically for AI job seekers, auto-generated to be accessible across devices. The project's steady growth suggests its usefulness in preparing candidates for the technical aspects of AI interviews with comprehensive and conveniently formatted resources.
gonemedia/aipointer (Growth Score: 20.03; Stars: 232) is an innovative tool that allows users to interact directly with their cursor context through voice or text queries, leveraging multiple AI providers for real-time responses on various operating systems. The rising star count and growth score reflect the community's interest in integrating advanced language models into everyday computing tasks.
rahilp/second-brain-cloudflare (Growth Score: 18.61; Stars: 89) provides a self-hosted memory layer that can be accessed across different AI tools, enabling users to store and recall information effortlessly on Cloudflare's free tier. The growth in stars and commits underscores the project’s appeal as a scalable solution for managing persistent data without relying on proprietary cloud services.
chiennv2000/orthrus (Growth Score: 17.20; Stars: 372) focuses on accelerating lossless LLM inference through dual-view diffusion decoding, aiming to improve the efficiency of large model deployments. The high star count and growth score suggest that developers are interested in optimizing AI performance for real-world applications where computational resources are limited.
okturro/asr-rescore-bench (Growth Score: 16.00; Stars: 154) offers a benchmarking framework to evaluate LLM-based ASR n-best rescoring strategies, including various linguistic models and prompt techniques. The steady growth in stars indicates the project's importance for researchers and developers looking to refine automatic speech recognition systems.
sitodowubb/spatial-vqa-bench (Growth Score: 15.20; Stars: 146) introduces a benchmark specifically designed to assess spatial visual reasoning capabilities of multimodal LLMs, focusing on their ability to interpret and respond accurately to queries involving spatial relationships in images or scenes. The growth score reflects the growing interest in evaluating AI's proficiency in understanding complex visual contexts.
ahammadmejbah/Awesome-Datasets-Hub (Growth Score: 14.54; Stars: 132) compiles a comprehensive list of datasets relevant to LLMs, covering areas such as medical AI, NLP, multimodal learning, and more. The increasing number of stars highlights the resource's value for researchers and developers seeking curated datasets for training and evaluating advanced language models.
Today's report showcases projects that range from enhancing user interactions with advanced AI capabilities to optimizing model performance and providing essential resources for research and development in the field of large language models.