---
license: mit
task_categories:
- text-generation
language:
- eo
- ga
- eu
- gl
- la
- cy
- so
- br
- sw
- is
- gd
- af
- am
- yi
- sa
- es
configs:
- config_name: eo
  data_files:
    - split: train
      path: "eo/train-*.parquet"
- config_name: ga
  data_files:
    - split: train
      path: "ga/train-*.parquet"
- config_name: eu
  data_files:
    - split: train
      path: "eu/train-*.parquet"
- config_name: gl
  data_files:
    - split: train
      path: "gl/train-*.parquet"
- config_name: la
  data_files:
    - split: train
      path: "la/train-*.parquet"
- config_name: cy
  data_files:
    - split: train
      path: "cy/train-*.parquet"
- config_name: so
  data_files:
    - split: train
      path: "so/train-*.parquet"
- config_name: br
  data_files:
    - split: train
      path: "br/train-*.parquet"
- config_name: sw
  data_files:
    - split: train
      path: "sw/train-*.parquet"
- config_name: is
  data_files:
    - split: train
      path: "is/train-*.parquet"
- config_name: gd
  data_files:
    - split: train
      path: "gd/train-*.parquet"
- config_name: af
  data_files:
    - split: train
      path: "af/train-*.parquet"
- config_name: am
  data_files:
    - split: train
      path: "am/train-*.parquet"
- config_name: yi
  data_files:
    - split: train
      path: "yi/train-*.parquet"
- config_name: sa
  data_files:
    - split: train
      path: "sa/train-*.parquet"
- config_name: es
  data_files:
    - split: train
      path: "es/train-*.parquet"
---

# CC100 Sharded Parquet Distribution (Tralalabs/CC100)

A high-performance, cleanly sharded, and memory-optimized multi-language distribution of the CC100 web-crawl dataset. Each language is stored inside its own sub-folder and isolated as an independent configuration block.

Stay tuned. More subsets are cooking in the kitchen.