fsdp repositories

Trending repositories tagged fsdp, ranked by trending score. Updated daily.

  1. 1
    NVIDIA-NeMo/labs-molt

    A scalable, agentic-first, and HuggingFace-native RL framework for research (9k lines).

    GitHub · Python · 1,401 stars · 154 forks · +38 stars in 24 hours · Trending score 3.11

  2. 2
    meta-pytorch/torchft

    Fault tolerance for PyTorch (HSDP, LocalSGD, DiLoCo, Streaming DiLoCo)

    GitHub · Python · 555 stars · 97 forks · +4 stars in 24 hours · Trending score 1.49

  3. 3
    sgl-project/SpecForge

    Train speculative decoding models effortlessly and port them smoothly to SGLang serving.

    GitHub · Python · 1,290 stars · 402 forks · +1 stars in 24 hours · Trending score 1.22

  4. 4
    YuvrajSingh-mist/smolcluster

    Educational distributed training and inference library for heterogeneous local hardware: Mac minis, Raspberry Pis and GPUs over plain Python sockets. FSDP, GRPO, parameter-server, expert and data parallelism.

    GitHub · Python · 97 stars · 6 forks · 0 stars in 24 hours · Trending score 0.89

  5. 5
    traceopt-ai/traceml

    Find why PyTorch training is slow, compare runs, and catch performance regressions in CI.

    GitHub · Python · 189 stars · 33 forks · 0 stars in 24 hours · Trending score 0.28

  6. 6
    lightseekorg/TorchSpec

    A PyTorch native library for training speculative decoding models

    GitHub · Python · 268 stars · 70 forks · 0 stars in 24 hours · Trending score 0.10

  7. 7
    Forge-researchlab/forge-kernels

    Forge kernels — fused Triton kernels for faster, leaner LLM fine-tuning. One-call patching into Hugging Face models via forge.patch(model), with FSDP2 multi-GPU support and a growing set of kernels and architectures. Every speedup backed by a committed benchmark. Apache-2.0.

    GitHub · Python · 11 stars · 5 forks · 0 stars in 24 hours · Trending score 0.00