Skip to content

Popular repositories Loading

  1. pegainfer pegainfer Public

    Pure Rust + CUDA LLM inference engine — no PyTorch, OpenAI-compatible, serves Qwen3 to Kimi-K2

    Rust 753 122

  2. kern kern Public

    Model-agnostic GPU runtime for LLM inference in Rust. Ship models as verified manifests + CUDA kernels — run standalone or drop into vLLM & SGLang.

    Rust 224 23

  3. kern-omni kern-omni Public

    OpenAI-compatible speech serving in Rust: streaming TTS and full-duplex voice over GPT-Live. 128 concurrent PersonaPlex sessions on one GPU.

    Rust 54 5

  4. kern-k31 kern-k31 Public

    Kimi-K3.1 Pro on the kern runtime: the manifest generator, kernels and pinned hashes for its P/D stack

    Cuda 8 1

  5. kern-rsi-demo kern-rsi-demo Public

    One GPU, a few coding agents, and a loop that makes Qwen3-4B cheaper to serve on kern, landing only measured, checked wins

    Python 8

  6. kern-k3-pp-prefill kern-k3-pp-prefill Public

    Kimi-K3 prefill as pipeline-parallel stages for kern: manifests, kernel sources, build recipes

    Cuda 6 1

Repositories

Showing 10 of 16 repositories
  • kern Public

    Model-agnostic GPU runtime for LLM inference in Rust. Ship models as verified manifests + CUDA kernels — run standalone or drop into vLLM & SGLang.

    pegainfer-project/kern's past year of commit activity
    Rust 224 Apache-2.0 23 30 12 Updated Nov 22, 2026
  • kern-inkling-2mini Public

    Inkling-2-Mini (inkling2_loop + DSpark) SM120 inference kernels and bundle for kern

    pegainfer-project/kern-inkling-2mini's past year of commit activity
    Python 2 Apache-2.0 1 0 0 Updated Nov 22, 2026
  • infront Public

    Implement one Scheduler, get an OpenAI-compatible inference server (vLLM Rust frontend)

    pegainfer-project/infront's past year of commit activity
    Rust 1 Apache-2.0 0 1 0 Updated Nov 21, 2026
  • pegainfer Public

    Pure Rust + CUDA LLM inference engine — no PyTorch, OpenAI-compatible, serves Qwen3 to Kimi-K2

    pegainfer-project/pegainfer's past year of commit activity
    Rust 753 Apache-2.0 122 36 (3 issues need help) 15 Updated Nov 21, 2026
  • kern-omni Public

    OpenAI-compatible speech serving in Rust: streaming TTS and full-duplex voice over GPT-Live. 128 concurrent PersonaPlex sessions on one GPU.

    pegainfer-project/kern-omni's past year of commit activity
    Rust 54 5 2 4 Updated Nov 20, 2026
  • kern-glm53 Public

    GLM-5.3 on the kern runtime: the manifest generator, kernels and pinned hashes for its pipeline-parallel prefill

    pegainfer-project/kern-glm53's past year of commit activity
    Cuda 0 Apache-2.0 0 1 0 Updated Nov 19, 2026
  • kern-rsi-demo Public

    One GPU, a few coding agents, and a loop that makes Qwen3-4B cheaper to serve on kern, landing only measured, checked wins

    pegainfer-project/kern-rsi-demo's past year of commit activity
    Python 8 Apache-2.0 0 0 0 Updated Nov 19, 2026
  • kern-k31 Public

    Kimi-K3.1 Pro on the kern runtime: the manifest generator, kernels and pinned hashes for its P/D stack

    pegainfer-project/kern-k31's past year of commit activity
    Cuda 8 Apache-2.0 1 0 0 Updated Nov 18, 2026
  • kern-qwen38 Public

    Qwen3.8-27B FP8 inference with native DSpine decoding for kern on SM120

    pegainfer-project/kern-qwen38's past year of commit activity
    Python 0 Apache-2.0 0 0 0 Updated Nov 18, 2026
  • kern-atlas Public

    Where GPU kernels already exist: one file each, for agents to find before writing another

    pegainfer-project/kern-atlas's past year of commit activity
    Python 0 0 0 0 Updated Nov 16, 2026

People

This organization has no public members. You must be a member to see who’s a part of this organization.