Meet the team that turns your ideas into work.

Discover EightBots
EIGHTBOTS
Contact salesSign up
AI provider

EleutherAI

EleutherAI develops language models that researchers can download and study. Its work includes the GPT-Neo lineage, Pythia training checkpoints and specialist models for Korean text and mathematics.

Explore models

Models & tools

  • A tiny six-layer model for studying the effect of training data.

    Launch date Research releaseDownloadable

  • A six-layer Pythia model that doubles the width of the 14M network.

    Launch date Research releaseDownloadable

  • A controlled suite of 6.9B models studying training-data filtering and safety.

    Launch date August 2025Research releaseDownloadable

  • A seven-billion-parameter Common Pile model trained for one trillion tokens.

    Launch date Research releaseDownloadable

  • A seven-billion-parameter Common Pile model trained for two trillion tokens.

    Launch date Research releaseDownloadable

  • A compact encoder-decoder model trained on two trillion Pile tokens.

    Launch date Research releaseDownloadable

  • A 783-million-parameter encoder-decoder model trained from scratch for text-to-text research.

    Launch date Research releaseDownloadable

  • A 2.85-billion-parameter text-to-text model with an open training history.

    Launch date Research releaseDownloadable

  • An eleven-billion-parameter encoder-decoder model trained from scratch.

    Launch date Research releaseDownloadable

  • A seven-billion-parameter RWKV chat model with recurrent inference.

    Launch date February 2024Research releaseDownloadable

  • A 34B mathematical language model adapted from Meta Code Llama.

    Launch date Research releaseDownloadable

  • A 7B mathematical language model adapted from Meta Code Llama.

    Launch date Research releaseDownloadable

  • A 14M Pythia checkpoint trained on deduplicated Pile text.

    Launch date July 2023Research releaseDownloadable

  • A 31M Pythia checkpoint trained on deduplicated Pile text.

    Launch date July 2023Research releaseDownloadable

  • A 24-layer Pythia network with a two-thousand-wide hidden state.

    Launch date Research releaseDownloadable

  • A 1.4B Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • The largest Pythia size, with 36 layers and open training checkpoints.

    Launch date Research releaseDownloadable

  • A 12B Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • A twelve-layer open model with checkpoints across its training run.

    Launch date Research releaseDownloadable

  • A 160M Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • A wide, sixteen-layer Pythia model trained with bfloat16 precision.

    Launch date Research releaseDownloadable

  • A 1B Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • A 32-layer Pythia model with about 2.52 billion non-embedding parameters.

    Launch date Research releaseDownloadable

  • A 2.8B Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • A 24-layer research model with a 1,024-wide hidden state.

    Launch date Research releaseDownloadable

  • A 410M Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • A 32-layer open model for research at a six-billion-parameter scale.

    Launch date Research releaseDownloadable

  • A 6.9B Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • The smallest size in the original eight-size Pythia research suite.

    Launch date Research releaseDownloadable

  • A 70M Pythia checkpoint trained on deduplicated Pile text.

    Launch date Research releaseDownloadable

  • A 20-billion-parameter Pile model from EleutherAI distributed training research.

    Launch date Research releaseDownloadable

  • A six-billion-parameter text model built with Mesh Transformer JAX.

    Launch date Research releaseDownloadable

  • An early open GPT-style model trained on 380 billion tokens.

    Launch date Research releaseDownloadable

  • The largest original GPT-Neo checkpoint, with 32 Transformer layers.

    Launch date Research releaseDownloadable

  • A compact English text model with local and global attention.

    Launch date not confirmedResearch releaseDownloadable

  • A chat adaptation of the Pile-trained Mamba 2.8B base.

    Launch date not confirmedResearch releaseDownloadable

  • A Mamba chat model whose base was trained on SlimPajama.

    Launch date not confirmedResearch releaseDownloadable

  • A preference-trained variant of the SlimPajama Hermes Mamba model.

    Launch date not confirmedResearch releaseDownloadable

  • A three-billion-parameter recurrent chat adaptation of RWKV v5.

    Launch date not confirmedResearch releaseDownloadable

  • An instruction-tuned Llemma adaptation with a nonstandard end token.

    Launch date not confirmedResearch releaseDownloadable

  • A compact Korean base model from the Polyglot project.

    Launch date not confirmedResearch releaseDownloadable

  • The largest documented Polyglot-Ko base model, trained on Korean text.

    Launch date not confirmedResearch releaseDownloadable

  • A Korean text-completion model with 32 Transformer layers.

    Launch date not confirmedResearch releaseDownloadable

  • A wider Korean base model with 28 layers and downloadable weights.

    Launch date not confirmedResearch releaseDownloadable

  • Language-model adaptations designed to separate stated answers from internal knowledge.

    Launch date not confirmedResearch releaseDownloadable