Adapt
Mini toolsChinchilla
Chinchilla is a 70B-parameter DeepMind language model trained on 1.3T tokens. It is best known for the Chinchilla scaling-law result, showing that for a fixed compute budget, smaller models trained on more data can outperform larger undertrained models, including Gopher, despite having far fewer parameters.
Text
Gen 2
Overview
Chinchilla is DeepMind’s 70B language model used to demonstrate compute-optimal scaling with more training data and fewer parameters than larger undertrained models.
About Google DeepMind
View Company ProfileTools using Chinchilla
No tools found for this model yet.
Last updated: June 2, 2026
