Skip to content
@infinigence

Infinigence

Popular repositories Loading

  1. Infini-Megrez Infini-Megrez Public

    337 20

  2. FlashOverlap FlashOverlap Public

    A lightweight design for computation-communication overlap.

    Python 249 18

  3. Infini-Megrez-Omni Infini-Megrez-Omni Public

    Python 243 9

  4. Semi-PD Semi-PD Public

    A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.

    Python 136 16

  5. FUSCO FUSCO Public

    High-performance distributed data shuffling (all-to-all) library for MoE training and inference

    Python 127 11

  6. pdd pdd Public

    95

Repositories

Showing 10 of 23 repositories
  • ApxInf Public
    infinigence/ApxInf's past year of commit activity
    C++ 81 Apache-2.0 23 7 22 Updated Oct 12, 2026
  • octollm Public

    An LLM API gateway framework

    infinigence/octollm's past year of commit activity
    Go 20 Apache-2.0 5 0 1 Updated Oct 9, 2026
  • Infini-Memory Public

    Infini-Memory: A maintainable, text-based persistent memory architecture that organizes LLM agent memory as topic-structured documents. paper: https://arxiv.org/abs/2606.10677

    infinigence/Infini-Memory's past year of commit activity
    Python 20 Apache-2.0 4 0 0 Updated Sep 1, 2026
  • ElasticMegatron Public

    An augmented Megatron-LM that supports online adjustment of GPU numbers and parallelization schemes

    infinigence/ElasticMegatron's past year of commit activity
    Python 6 Apache-2.0 0 0 3 Updated Aug 10, 2026
  • pdd Public
    infinigence/pdd's past year of commit activity
    95 0 1 0 Updated Jul 17, 2026
  • tokenestimate Public

    A fast Go library for estimating token counts without actual tokenization.

    infinigence/tokenestimate's past year of commit activity
    Go 2 0 0 0 Updated Mar 27, 2026
  • FUSCO Public

    High-performance distributed data shuffling (all-to-all) library for MoE training and inference

    infinigence/FUSCO's past year of commit activity
    Python 127 Apache-2.0 11 1 0 Updated Mar 7, 2026
  • FlashOverlap Public

    A lightweight design for computation-communication overlap.

    infinigence/FlashOverlap's past year of commit activity
    Python 249 Apache-2.0 18 9 2 Updated Jan 20, 2026
  • Semi-PD Public

    A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.

    infinigence/Semi-PD's past year of commit activity
    Python 136 Apache-2.0 16 7 1 Updated Dec 25, 2025
  • infinigence/HamiltonAttention's past year of commit activity
    Python 46 3 0 0 Updated Oct 15, 2025

Top languages

Loading…

Most used topics

Loading…