Evaluation datasets made by TNSA
TNSA
company
Verified
AI & ML interests
Building what the world needs.
Recent Activity
The NGen-4-OW models are built on top of distillation data of NGen-4 Pro Model by TNSA running on NGen-4-OW Architecture.
TNSA AI's NGen 3 Model Squad the Architecture is not Open Source, only the weights are Open please visit, https://tnsaai.com to read System Card
Benchmark dataset for Benching
Best Reasoning Dataset in One Place
All Models from the NGen2 Model Family, Pre-Trained for Token-Prediction, are experimental and not made for Production use-case. Check out NGen-4-OW.
This collection gathers both external (copied) and TNSA-owned datasets to streamline data imports from a Single Org.
TNSA's OpenWeight Model distilled from NGen4 Foundational Models "Note : We recommend you to use the new NGen-4-OW Family instead of this"
TNSA's NGen4 Family this Collection Includes Experimental Model Builds, these Models are the experimental Architectures and Experiments done.
Datasets for Training of High-Level AI Systems, and few Extractions from them.
Datasets for Training Models
Evaluation datasets made by TNSA
This collection gathers both external (copied) and TNSA-owned datasets to streamline data imports from a Single Org.
The NGen-4-OW models are built on top of distillation data of NGen-4 Pro Model by TNSA running on NGen-4-OW Architecture.
TNSA's OpenWeight Model distilled from NGen4 Foundational Models "Note : We recommend you to use the new NGen-4-OW Family instead of this"
TNSA AI's NGen 3 Model Squad the Architecture is not Open Source, only the weights are Open please visit, https://tnsaai.com to read System Card
TNSA's NGen4 Family this Collection Includes Experimental Model Builds, these Models are the experimental Architectures and Experiments done.
Benchmark dataset for Benching
Datasets for Training of High-Level AI Systems, and few Extractions from them.
Best Reasoning Dataset in One Place
Datasets for Training Models
All Models from the NGen2 Model Family, Pre-Trained for Token-Prediction, are experimental and not made for Production use-case. Check out NGen-4-OW.