Today's LLM & Language Models: Fastest-Growing Projects — April 22, 2026
Today's the LLM & Language Models space, we're seeing a surge in interest around personal knowledge bases and tools that make language models more accessible and user-friendly. The top-growing repositories are focused on creating structured wikis from unorganized data, removing AI "slop" from LLM outputs, and providing educational resources for understanding LLM internals.
Arman-bd's Guppylm is a ~9M parameter LLM that talks like a small fish, with a Growth Score of 80.48 and 2,975 stars. Its unique personality has captured the attention of many, making it one of the fastest-growing repositories in this category.
Hexiecs' talk-normal repository boasts a Growth Score of 65.43 and 1,379 stars. This system prompt removes AI "slop" from LLM outputs, making them sound more like normal human conversations, which is likely why it's gaining popularity among developers looking to refine their language models.
Sdyckjq-lab's llm-wiki-skill has a Growth Score of 56.24 and 967 stars. This personal knowledge base construction skill supports multiple platforms and is based on Karpathy's LLM Wiki method, which may explain its growth as more developers explore this approach to organizing their knowledge.
AmitShekhariitbhu's llm-internals repository has a Growth Score of 40.40 and 604 stars. This educational resource teaches LLM internals step-by-step, from tokenization to attention to inference optimization, likely appealing to developers seeking to deepen their understanding of language models.
Kessler's Gemma-Gem has a Growth Score of 35.62 and 815 stars. This tool runs Google's Gemma 4 model entirely on-device via WebGPU, eliminating the need for API keys, cloud services, or data leaving one's machine, which is likely driving its growth among developers prioritizing security and efficiency.
Xoai's sage-wiki boasts a Growth Score of 35.25 and 447 stars. This LLM-compiled personal knowledge base takes in papers, articles, and notes, compiling them into a structured, interlinked wiki with extracted concepts and cross-references – making it an attractive tool for knowledge management.
VectifyAI's OpenKB has a Growth Score of 31.47 and 461 stars. As an open LLM knowledge base, this repository is likely growing due to the increasing interest in collaborative, community-driven approaches to language model development.
Lucasastorian's llmwiki boasts a Growth Score of 30.56 and 596 stars. This open-source implementation of Karpathy's LLM Wiki allows users to upload documents and connect their Claude account via MCP, generating a wiki – making it an appealing option for those looking to leverage this popular approach.
Atomicmemory's llm-wiki-compiler has a Growth Score of 29.74 and 672 stars. This knowledge compiler takes raw sources and outputs an interlinked wiki, inspired by Karpathy's LLM Wiki pattern, which is likely driving its growth among developers seeking structured knowledge bases.
Pratiyush's llm-wiki repository has a Growth Score of 28.21 and 139 stars. By providing an LLM-powered knowledge base from various AI tool sessions, this tool is growing in popularity as more developers explore the potential of language models for personal knowledge management.
Arman-bd's Guppylm is a ~9M parameter LLM that talks like a small fish, with a Growth Score of 80.48 and 2,975 stars. Its unique personality has captured the attention of many, making it one of the fastest-growing repositories in this category.
Hexiecs' talk-normal repository boasts a Growth Score of 65.43 and 1,379 stars. This system prompt removes AI "slop" from LLM outputs, making them sound more like normal human conversations, which is likely why it's gaining popularity among developers looking to refine their language models.
Sdyckjq-lab's llm-wiki-skill has a Growth Score of 56.24 and 967 stars. This personal knowledge base construction skill supports multiple platforms and is based on Karpathy's LLM Wiki method, which may explain its growth as more developers explore this approach to organizing their knowledge.
AmitShekhariitbhu's llm-internals repository has a Growth Score of 40.40 and 604 stars. This educational resource teaches LLM internals step-by-step, from tokenization to attention to inference optimization, likely appealing to developers seeking to deepen their understanding of language models.
Kessler's Gemma-Gem has a Growth Score of 35.62 and 815 stars. This tool runs Google's Gemma 4 model entirely on-device via WebGPU, eliminating the need for API keys, cloud services, or data leaving one's machine, which is likely driving its growth among developers prioritizing security and efficiency.
Xoai's sage-wiki boasts a Growth Score of 35.25 and 447 stars. This LLM-compiled personal knowledge base takes in papers, articles, and notes, compiling them into a structured, interlinked wiki with extracted concepts and cross-references – making it an attractive tool for knowledge management.
VectifyAI's OpenKB has a Growth Score of 31.47 and 461 stars. As an open LLM knowledge base, this repository is likely growing due to the increasing interest in collaborative, community-driven approaches to language model development.
Lucasastorian's llmwiki boasts a Growth Score of 30.56 and 596 stars. This open-source implementation of Karpathy's LLM Wiki allows users to upload documents and connect their Claude account via MCP, generating a wiki – making it an appealing option for those looking to leverage this popular approach.
Atomicmemory's llm-wiki-compiler has a Growth Score of 29.74 and 672 stars. This knowledge compiler takes raw sources and outputs an interlinked wiki, inspired by Karpathy's LLM Wiki pattern, which is likely driving its growth among developers seeking structured knowledge bases.
Pratiyush's llm-wiki repository has a Growth Score of 28.21 and 139 stars. By providing an LLM-powered knowledge base from various AI tool sessions, this tool is growing in popularity as more developers explore the potential of language models for personal knowledge management.