Skip to content

@neuralmagic Neural Magic

Change the repository type filter

All

    Repositories list

    93 repositories

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •16k•17•0•38•Updated Apr 23, 2026Apr 23, 2026

    • nm-vllm-omni-ent

      Public
      A framework for efficient model inference with omni-modality models
      Python
      •
      Apache License 2.0
      •815•2•0•1•Updated Apr 23, 2026Apr 23, 2026

    • GuardBench

      Public
      A Python library for guardrail models evaluation with vLLM support.
      Python
      •
      European Union Public License 1.2
      •10•0•0•15•Updated Apr 21, 2026Apr 21, 2026

    • every_eval_ever

      Public
      Every Eval Ever is a shared schema and crowdsourced eval database. It defines a standardized metadata format for storing AI evaluation results — from leaderboar…
      Python
      •
      MIT License
      •29•0•0•6•Updated Apr 21, 2026Apr 21, 2026

    • axolotl

      Public
      Go ahead and axolotl questions
      Python
      •
      Apache License 2.0
      •1.3k•0•0•6•Updated Apr 21, 2026Apr 21, 2026

    • model-validation-configs

      Public
      1•2•1•4•Updated Apr 21, 2026Apr 21, 2026

    • nm-actions

      Public
      Neural Magic GHA
      Python
      •
      Apache License 2.0
      •0•0•0•5•Updated Apr 20, 2026Apr 20, 2026

    • lighteval

      Public
      Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends
      Python
      •
      MIT License
      •452•0•0•1•Updated Apr 18, 2026Apr 18, 2026

    • nyann-bench

      Public
      Go
      •
      Apache License 2.0
      •0•0•0•4•Updated Apr 17, 2026Apr 17, 2026

    • research

      Public
      Repository to enable research flows
      Python
      •0•3•0•3•Updated Apr 17, 2026Apr 17, 2026

    • flash-attention

      Public
      Fast and memory-efficient exact attention
      C++
      •
      BSD 3-Clause "New" or "Revised" License
      •2.6k•0•0•0•Updated Apr 16, 2026Apr 16, 2026

    • transformers-gdm

      Public
      🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference an…
      Python
      •
      Apache License 2.0
      •33k•0•0•0•Updated Apr 15, 2026Apr 15, 2026

    • lm-evaluation-harness

      Public
      A framework for few-shot evaluation of language models.
      Python
      •
      MIT License
      •3.2k•5•0•1•Updated Apr 14, 2026Apr 14, 2026

    • trending_benchmarks

      Public
      Tool to scrape benchmarks used most commonly in recent popular open source models
      Python
      •
      MIT License
      •1•0•0•0•Updated Apr 11, 2026Apr 11, 2026

    • vllm-beamsearch-plugin

      Public
      Beam search scheduler plugin for vLLM v1 with CoW block table forking
      Python
      •0•0•0•0•Updated Apr 9, 2026Apr 9, 2026

    • SWE-bench

      Public
      SWE-bench: Can Language Models Resolve Real-world Github Issues?
      Python
      •
      MIT License
      •836•0•0•0•Updated Apr 8, 2026Apr 8, 2026

    • sglang

      Public
      SGLang is a fast serving framework for large language models and vision language models.
      Python
      •
      Apache License 2.0
      •5.5k•1•0•3•Updated Apr 8, 2026Apr 8, 2026

    • lmms-eval

      Public
      Accelerating the development of large multimodal models (LMMs) with one-click evaluation module - lmms-eval.
      Python
      •
      Other
      •566•0•0•12•Updated Mar 12, 2026Mar 12, 2026

    • DeepEP

      Public
      DeepEP: an efficient expert-parallel communication library
      Cuda
      •
      MIT License
      •1.2k•1•0•0•Updated Mar 11, 2026Mar 11, 2026

    • mini-swe-agent

      Public
      The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-b…
      Python
      •
      MIT License
      •549•0•0•0•Updated Mar 10, 2026Mar 10, 2026

    • vllm-fork

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •16k•0•0•1•Updated Mar 5, 2026Mar 5, 2026

    • tpu-inference

      Public
      TPU inference for vLLM, with unified JAX and PyTorch support.
      Python
      •
      Apache License 2.0
      •169•0•0•0•Updated Mar 5, 2026Mar 5, 2026

    • arena-hard-auto

      Public
      Arena-Hard-Auto: An automatic LLM benchmark.
      Python
      •
      Apache License 2.0
      •150•0•0•3•Updated Mar 3, 2026Mar 3, 2026

    • pytorch

      Public
      Tensors and Dynamic neural networks in Python with strong GPU acceleration
      Python
      •
      Other
      •28k•1•0•6•Updated Feb 11, 2026Feb 11, 2026

    • nm-vllm

      Public archive
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Other
      •16k•266•0•0•Updated Dec 4, 2025Dec 4, 2025

    • speculators-research

      Public
      Python
      •1•0•0•1•Updated Nov 13, 2025Nov 13, 2025

    • opendatahub-operator

      Public
      Open Data Hub operator to manage ODH component integrations
      Go
      •
      Apache License 2.0
      •254•0•0•0•Updated Nov 12, 2025Nov 12, 2025

    • DeepEP-test

      Public
      DeepEP: an efficient expert-parallel communication library
      Cuda
      •
      MIT License
      •1.2k•0•0•0•Updated Sep 26, 2025Sep 26, 2025

    • pydantic-regmix

      Public
      Common mixins, registries, and utilities with native support for Pydantic used across popular repos such as GuideLLM and Speculators
      Apache License 2.0
      •0•0•0•0•Updated Sep 17, 2025Sep 17, 2025

    • upstream-transformers

      Public
      🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
      Python
      •
      Apache License 2.0
      •33k•2•0•0•Updated Sep 12, 2025Sep 12, 2025
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.