Skip to content

Popular repositories Loading

  1. exllamav2 exllamav2 Public

    A fast inference library for running LLMs locally on modern consumer-class GPUs

    Python 4.2k 321

  2. exui exui Public

    Web UI for ExLlamaV2

    JavaScript 505 48

  3. exllamav3 exllamav3 Public

    An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs

    Python 449 36

Repositories

Showing 3 of 3 repositories
  • exllamav3 Public

    An optimized quantization and inference library for running LLMs locally on modern consumer-class GPUs

    turboderp-org/exllamav3’s past year of commit activity
    Python 449 MIT 36 25 1 Updated Jul 20, 2025
  • exllamav2 Public

    A fast inference library for running LLMs locally on modern consumer-class GPUs

    turboderp-org/exllamav2’s past year of commit activity
    Python 4,244 MIT 321 128 20 Updated Jul 14, 2025
  • exui Public

    Web UI for ExLlamaV2

    turboderp-org/exui’s past year of commit activity
    JavaScript 505 MIT 48 33 3 Updated Feb 6, 2025

Most used topics

Loading…