Sakana AI
On a quest to create a new kind of foundation model based on nature-inspired intelligence.
Tokyo
Pinned Repositories
AI-Scientist
The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑🔬
AI-Scientist-v2
The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
asal
Automating the Search for Artificial Life with Foundation Models!
continuous-thought-machines
Continuous Thought Machines, because thought takes time and reasoning is a process.
evo-memory
Code to train and evaluate Neural Attention Memory Models to obtain universally-applicable memory systems for transformers.
evolutionary-model-merge
Official repository of Evolutionary Optimization of Model Merging Recipes
RLT
Training teachers with reinforcement learning able to make LLMs learn how to reason for test time scaling.
self-adaptive-llms
A Self-adaptation Framework🐙 that adapts LLMs for unseen tasks in real-time!
text-to-lora
Hypernetworks that adapt LLMs for specific benchmark tasks using only textual task description as the input
treequest
A Tree Search Library with Flexible API for LLM Inference-Time Scaling
Sakana AI's Repositories
SakanaAI/AI-Scientist
The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑🔬
SakanaAI/AI-Scientist-v2
The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
SakanaAI/continuous-thought-machines
Continuous Thought Machines, because thought takes time and reasoning is a process.
SakanaAI/evolutionary-model-merge
Official repository of Evolutionary Optimization of Model Merging Recipes
SakanaAI/self-adaptive-llms
A Self-adaptation Framework🐙 that adapts LLMs for unseen tasks in real-time!
SakanaAI/text-to-lora
Hypernetworks that adapt LLMs for specific benchmark tasks using only textual task description as the input
SakanaAI/treequest
A Tree Search Library with Flexible API for LLM Inference-Time Scaling
SakanaAI/asal
Automating the Search for Artificial Life with Foundation Models!
SakanaAI/RLT
Training teachers with reinforcement learning able to make LLMs learn how to reason for test time scaling.
SakanaAI/evo-memory
Code to train and evaluate Neural Attention Memory Models to obtain universally-applicable memory systems for transformers.
SakanaAI/AI-Scientist-ICLR2025-Workshop-Experiment
SakanaAI/DiscoPOP
Code for Discovering Preference Optimization Algorithms with and for Large Language Models
SakanaAI/natural_niches
The code repository of the paper: Competition and Attraction Improve Model Fusion
SakanaAI/TinySwallow-ChatUI
Browser-based chat UI for TinySwallow-1.5B that runs without API calls.
SakanaAI/ALE-Bench
The official repository of ALE-Bench
SakanaAI/TAID
Official implementation of "TAID: Temporally Adaptive Interpolated Distillation for Efficient Knowledge Transfer in Language Models"
SakanaAI/Sudoku-Bench
An AI benchmark for creative, human-like problem solving using Sudoku variants
SakanaAI/ab-mcts-arc2
SakanaAI/TinySwallow-ChatUI-Local
Python-based chat demo for TinySwallow-1.5B that works completely offline
SakanaAI/CycleQD
CycleQD is a framework for parameter space model merging.
SakanaAI/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
SakanaAI/EDINET-Bench
Evaluating the performance of LLMs on Japanese challenging financial tasks.
SakanaAI/TransEvalnia
Reasoning-based Evaluation and Ranking of Translations.
SakanaAI/L2D
Large language models to diffusion finetuning code
SakanaAI/edinet2dataset
edinet2dataset is a tool to construct financial dataset using EDINET.
SakanaAI/nca-alife
Learning Neural Cellular Automata that produce Open-Ended Alife!
SakanaAI/BALROG
Benchmarking Agentic LLM and VLM Reasoning On Games
SakanaAI/petri-dish-nca