Today's AI Frameworks & SDKs: Fastest-Growing Projects — May 17, 2026
This week, the AI Frameworks & SDKs space has seen a surge of innovative projects aimed at enhancing performance and usability for large language models (LLMs) and multi-agent systems. The standout tools are those that offer speed enhancements or unique orchestration capabilities, with developers showing strong interest in frameworks that can optimize resource allocation and streamline complex workflows.
lightseekorg/tokenspeed
TokenSpeed is a high-performance inference engine designed to process LLMs at near-light speeds. Its impressive growth score of 86.95 reflects the community's enthusiasm for its potential to significantly accelerate AI applications, with over 1,031 stars indicating strong developer support.
2508965-ship-it/harmonist-orchestral
Harmonist Orchestral is a multi-agent orchestration engine that enables the construction and deployment of AI swarms. The project's growth score of 71.00 and 420 stars highlight its potential to simplify complex AI coordination tasks, making it an attractive tool for developers looking to build sophisticated AI systems.
simonlin1212/TradingAgents-astock
A multi-agent investment research framework tailored specifically for the A-share market in China, TradingAgents-Astock integrates a variety of data sources and employs seven AI analysts to debate decisions. With a growth score of 69.00 and 276 stars, its specialized approach to financial analysis is gaining traction among those interested in leveraging AI for trading strategies.
1bananachicken/MaaNTE
MaaNTE is described as an assistant driven by the MAAFramework, designed primarily for automation tasks within a specific game context. Its strong growth score of 54.45 and 1,324 stars suggest that it resonates with developers seeking to automate repetitive or complex in-game activities.
noonghunna/club-3090
This repository offers community-driven recipes for serving large language models on RTX 3090 GPUs, supporting multiple engines and model configurations. Its growth score of 44.97 and 962 stars indicate that it is valuable to those optimizing LLM performance on high-end hardware.
TheRunicDev/MaaNTE
Another project leveraging the MAAFramework for automation tasks in a game environment, this version focuses specifically on features like auto-fishing and daily reward claiming. With a growth score of 40.90 and 394 stars, it appeals to users looking for comprehensive automation solutions.
oritera/Cairn
Cairn is a general-purpose state-space search engine that has been validated through autonomous penetration testing. Its growth score of 30.95 and 1,019 stars reflect the interest in its versatile approach to solving complex computational problems across different domains.
alash3al/stash
Stash provides a persistent memory layer for AI agents, storing episodes, facts, and working context within Postgres databases. Its growth score of 29.76 and 694 stars highlight the demand for robust storage solutions that enhance the functionality of AI systems by maintaining stateful information.
Ontos-AI/knowhere
Knowhere is a tool designed to extract, parse, and structure data chunks suitable for AI agents and retrieval-augmented generation (RAG) applications. With a growth score of 26.09 and 215 stars, it addresses the need for efficient data processing in complex AI workflows.
pouyahasanamreji/continuum
Continuum offers shared memory and orchestration capabilities designed to facilitate collaboration among coding agents. Its growth score of 21.60 and 27 stars indicate its relevance to developers seeking streamlined integration and persistent vector memory for their projects.
lightseekorg/tokenspeed
TokenSpeed is a high-performance inference engine designed to process LLMs at near-light speeds. Its impressive growth score of 86.95 reflects the community's enthusiasm for its potential to significantly accelerate AI applications, with over 1,031 stars indicating strong developer support.
2508965-ship-it/harmonist-orchestral
Harmonist Orchestral is a multi-agent orchestration engine that enables the construction and deployment of AI swarms. The project's growth score of 71.00 and 420 stars highlight its potential to simplify complex AI coordination tasks, making it an attractive tool for developers looking to build sophisticated AI systems.
simonlin1212/TradingAgents-astock
A multi-agent investment research framework tailored specifically for the A-share market in China, TradingAgents-Astock integrates a variety of data sources and employs seven AI analysts to debate decisions. With a growth score of 69.00 and 276 stars, its specialized approach to financial analysis is gaining traction among those interested in leveraging AI for trading strategies.
1bananachicken/MaaNTE
MaaNTE is described as an assistant driven by the MAAFramework, designed primarily for automation tasks within a specific game context. Its strong growth score of 54.45 and 1,324 stars suggest that it resonates with developers seeking to automate repetitive or complex in-game activities.
noonghunna/club-3090
This repository offers community-driven recipes for serving large language models on RTX 3090 GPUs, supporting multiple engines and model configurations. Its growth score of 44.97 and 962 stars indicate that it is valuable to those optimizing LLM performance on high-end hardware.
TheRunicDev/MaaNTE
Another project leveraging the MAAFramework for automation tasks in a game environment, this version focuses specifically on features like auto-fishing and daily reward claiming. With a growth score of 40.90 and 394 stars, it appeals to users looking for comprehensive automation solutions.
oritera/Cairn
Cairn is a general-purpose state-space search engine that has been validated through autonomous penetration testing. Its growth score of 30.95 and 1,019 stars reflect the interest in its versatile approach to solving complex computational problems across different domains.
alash3al/stash
Stash provides a persistent memory layer for AI agents, storing episodes, facts, and working context within Postgres databases. Its growth score of 29.76 and 694 stars highlight the demand for robust storage solutions that enhance the functionality of AI systems by maintaining stateful information.
Ontos-AI/knowhere
Knowhere is a tool designed to extract, parse, and structure data chunks suitable for AI agents and retrieval-augmented generation (RAG) applications. With a growth score of 26.09 and 215 stars, it addresses the need for efficient data processing in complex AI workflows.
pouyahasanamreji/continuum
Continuum offers shared memory and orchestration capabilities designed to facilitate collaboration among coding agents. Its growth score of 21.60 and 27 stars indicate its relevance to developers seeking streamlined integration and persistent vector memory for their projects.