Skip to content

@neuralmagic Neural Magic

Change the repository type filter

All

    Repositories list

    99 repositories

    • vllm-vcr

      Public
      vLLM inference simulator and replay machine
      rustinferencevllm
      rustinferencevllm
      Rust
      •
      Apache License 2.0
      •1•4•12•8•Updated Jul 19, 2026Jul 19, 2026

    • axolotl

      Public
      Go ahead and axolotl questions
      Python
      •
      Apache License 2.0
      •1.4k•0•0•6•Updated Jul 19, 2026Jul 19, 2026

    • DeepGEMM

      Public
      DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
      Cuda
      •
      MIT License
      •1.1k•0•0•2•Updated Jul 18, 2026Jul 18, 2026

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •20k•17•0•37•Updated Jul 18, 2026Jul 18, 2026

    • gorilla

      Public
      Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)
      Python
      •
      Apache License 2.0
      •1.4k•0•0•15•Updated Jul 17, 2026Jul 17, 2026

    • LMCache

      Public
      Redis for LLMs
      Python
      •
      Apache License 2.0
      •1.6k•0•0•1•Updated Jul 17, 2026Jul 17, 2026

    • tokenspeed

      Public
      TokenSpeed is a speed-of-light LLM inference engine.
      Python
      •
      MIT License
      •193•0•0•2•Updated Jul 17, 2026Jul 17, 2026

    • speculators

      Public
      A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM
      Python
      •
      Apache License 2.0
      •160•0•0•1•Updated Jul 17, 2026Jul 17, 2026

    • model-validation-configs

      Public
      2•2•1•4•Updated Jul 17, 2026Jul 17, 2026

    • nyann-bench

      Public
      Go
      •
      Apache License 2.0
      •1•0•0•3•Updated Jul 15, 2026Jul 15, 2026

    • research

      Public
      Repository to enable research flows
      Python
      •1•3•0•3•Updated Jul 14, 2026Jul 14, 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
      •42•0•0•17•Updated Jul 10, 2026Jul 10, 2026

    • lm-evaluation-harness

      Public
      A framework for few-shot evaluation of language models.
      Python
      •
      MIT License
      •3.4k•6•0•4•Updated Jul 10, 2026Jul 10, 2026

    • DeepEP

      Public
      DeepEP: an efficient expert-parallel communication library
      Cuda
      •
      MIT License
      •1.3k•1•0•0•Updated Jul 10, 2026Jul 10, 2026

    • relay-testbed

      Public
      Asynchronous message-relay pipeline used as an integration-test fixture for internal tooling.
      Rust
      •
      Apache License 2.0
      •1•0•1•0•Updated Jul 7, 2026Jul 7, 2026

    • opendatahub-operator

      Public
      Open Data Hub operator to manage ODH component integrations
      Go
      •
      Apache License 2.0
      •294•0•0•1•Updated Jul 3, 2026Jul 3, 2026

    • nm-vllm-omni-ent

      Public
      A framework for efficient model inference with omni-modality models
      Python
      •
      Apache License 2.0
      •1.3k•3•0•0•Updated Jun 30, 2026Jun 30, 2026

    • vllm-openshift-recipes

      Public
      0•0•0•0•Updated Jun 25, 2026Jun 25, 2026

    • vllm-beamsearch-plugin

      Public
      Beam search scheduler plugin for vLLM v1 with CoW block table forking
      Python
      •0•0•1•0•Updated Jun 23, 2026Jun 23, 2026

    • lighteval

      Public
      Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends
      Python
      •
      MIT License
      •513•0•0•1•Updated Jun 18, 2026Jun 18, 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 May 29, 2026May 29, 2026

    • GuardBench

      Public
      A Python library for guardrail models evaluation with vLLM support.
      Python
      •
      European Union Public License 1.2
      •11•0•0•16•Updated May 11, 2026May 11, 2026

    • nm-actions

      Public
      Neural Magic GHA
      Python
      •
      Apache License 2.0
      •0•0•0•4•Updated Apr 20, 2026Apr 20, 2026

    • flash-attention

      Public
      Fast and memory-efficient exact attention
      C++
      •
      BSD 3-Clause "New" or "Revised" License
      •2.9k•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
      •34k•0•0•0•Updated Apr 15, 2026Apr 15, 2026

    • SWE-bench

      Public
      SWE-bench: Can Language Models Resolve Real-world Github Issues?
      Python
      •
      MIT License
      •924•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
      •7.3k•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
      •623•0•0•12•Updated Mar 12, 2026Mar 12, 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
      •820•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
      •20k•0•0•1•Updated Mar 5, 2026Mar 5, 2026
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.