List your company

ResearchFormerly WizWand

LLM

Benchmarks

TaskDatasetSOTA resultTrendResults

Binary Inconsistency Detection

LLM

70.27Accuracy

10

Robust Steganography

LLM Generative Text

84.08Embedding Capacity (bits / 1k tokens)

5

LLM

0.332F1 Score

5

Language Modeling

LLM (val)

1.336Loss

4

LLM-329M

205Peak Performance (FP4/FP8)

1