PullRepo

Daily radar for the fastest-growing AI tools & repos

Today's AI Research: Fastest-Growing Projects — May 10, 2026

Today's AI Research, we're seeing a surge of interest in multimodal intelligence, language models, and autonomous agents. Researchers are pushing the boundaries of what's possible with large language models (LLMs), exploring new applications in finance, healthcare, and beyond. Meanwhile, efforts to improve the consistency and reliability of these models are gaining traction.

Trading-agents from lukiIabs takes the top spot this week with a growth score of 53.89 and 232 stars. This repository is building a multi-agent trading platform that leverages LLMs for stock and cryptocurrency trading, sentiment analysis, and quantitative algo trading – its popularity reflects the growing interest in applying AI to finance. With 15 commits over the past month, this project is rapidly evolving to meet the needs of researchers and practitioners.

Fkyah3's opencode-yg has a growth score of 17.92 and 36 stars, but what really stands out is its impressive 100 commits over the past month. This research fork demonstrates Language Anchoring, which enables LLMs to think consistently in a specific language – with verified results showing 95%+ Chinese thinking compliance, it's no wonder this project is gaining attention.

Matrix-agent's awesome-agentic-world-modeling boasts an impressive 194 stars and a growth score of 8.78. This repository provides a comprehensive overview of Agentic World Modeling, covering foundations, capabilities, laws, and beyond – its popularity reflects the growing interest in understanding how agents interact with their environments. With 12 commits over the past month, this resource is continually being updated to reflect the latest developments in the field.

AutoMedBench's eponymous repository has a growth score of 8.44 and 25 stars, but what's notable is its 46 commits over the past month. This Medical AutoResearch Benchmark for Autonomous AI Agents aims to provide a standardized evaluation framework for medical applications – as interest in healthcare AI grows, so too does the need for robust benchmarks like this one.

Thunlp's OPD has an impressive 293 stars and a growth score of 7.54. This repository is dedicated to the paper "Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe" – its popularity reflects the ongoing efforts to improve LLMs through distillation techniques. With 17 commits over the past month, this project continues to evolve as new insights emerge.

XIAO4579's PRISM boasts a growth score of 5.88 and 68 stars. This repository explores pre-alignment via black-box on-policy distillation for multimodal RL – its growing popularity reflects the need for more effective methods in this area. With 19 commits over the past month, this project is actively being developed to address the challenges of multimodal learning.

Gameworld-project's gameworld has a growth score of 5.12 and 172 stars. This repository aims to provide standardized and verifiable evaluation of multimodal game agents – its popularity reflects the growing interest in applying AI to games and simulations. With 11 commits over the past month, this project continues to refine its approach.

Hedlen's Awesome-Multimodal-Intelligence has a growth score of 3.68 and 40 stars. This curated collection covers VLMs, VLAs, World Models, and embodied AI – its popularity reflects the need for centralized resources in this rapidly evolving field. With 8 commits over the past month, this repository remains up-to-date with the latest developments.

AMAP-ML's DCW has a growth score of 3.54 and 115 stars. This repository explores the SNR-t bias of diffusion probabilistic models – its growing popularity reflects the ongoing efforts to understand and improve these models. With 7 commits over the past month, this project continues to refine its insights.

Yovecent's UDM-GRPO has a growth score of 3.26 and 23 stars, but what stands out is its impressive 25 commits over the past month. This repository presents stable and efficient group relative policy optimization for uniform discrete diffusion models – as interest in these models grows, so too does the need for effective optimization techniques like this one.
Back to all reports