Open-source llm-inference software
3358 open-source libraries, tools and applications tagged llm-inference, ranked by how closely llm-inference is their primary purpose, then by adoption and maintenance health. Each entry links to a page with sources, use cases and when to choose or avoid it.
3358 products, primary matches first, then adoption-weighted; health v2 shown.
| Product | Health v2 | Stars | Maturity |
|---|---|---|---|
| HKUDS/AutoAgent AutoAgent is a fully-automated, zero-code framework for creating and deploying LLM agents using natural language alone. It acts as an auton… | 39 | 9806 | active |
| tmc/langchaingo Langchaingo is the Go language port of LangChain, providing composable building blocks for writing LLM-powered applications in Go. It offer… | 87 | 9712 | active |
| mshumer/gpt-prompt-engineer A Jupyter Notebook-based tool that automatically generates, tests, and ranks LLM prompts for a given task using GPT-4, GPT-3.5-Turbo, or Cl… | 47 | 9681 | active |
| xorbitsai/inference Xinference is an open-source model serving platform for deploying LLMs, embedding, speech, image, and multimodal models via a unified OpenA… | 93 | 9616 | active |
| Mobilerun Mobilerun is an open-source Python framework for controlling Android and iOS devices with LLM-powered agents using natural language command… | 82 | 9613 | active |
| zhouxiaoka/autoclip AutoClip is a self-hosted AI-powered video clipping system that downloads videos from YouTube and Bilibili, uses LLM-based content analysis… | 86 | 9577 | active |
| Spring AI Spring AI is an application framework for AI engineering that provides Spring-friendly, portable abstractions for integrating AI models int… | 94 | 9542 | active |
| fastrepl/anarlog Anarlog is an open-source, local-first AI meeting notepad and Granola alternative that records device audio without a bot joining the call,… | 85 | 9527 | active |
| BAML BAML is a domain-specific programming language for building LLM agents, with a Rust-like type system, typed errors, built-in tests/evals, a… | 92 | 9423 | active |
| oumi-ai/oumi Oumi is an open-source Python framework and platform for the end-to-end lifecycle of open-weight LLMs: data synthesis, fine-tuning (SFT, Lo… | 91 | 9407 | active |
| Future-House/paper-qa PaperQA2 is a Python library and CLI for high-accuracy retrieval augmented generation (RAG) over PDFs, text, Office documents, and source c… | 95 | 9365 | active |
| LearningCircuit/local-deep-research A self-hosted AI research assistant that performs deep, iterative multi-source research using local or cloud LLMs (Ollama, llama.cpp, OpenA… | 82 | 9359 | active |
| alvinunreal/oh-my-opencode-slim A lightweight agent orchestration plugin for OpenCode that delegates tasks to a team of seven specialized sub-agents (Orchestrator, Explore… | 78 | 9344 | active |
| shaxiu/XianyuAutoAgent An AI-powered customer service bot for the Xianyu (Goofish) secondhand marketplace that provides 24/7 automated message handling. It uses L… | 56 | 9275 | active |
| MervinPraison/PraisonAI PraisonAI is a Python multi-agent framework for building and deploying autonomous, self-improving AI agents with built-in memory, RAG, and … | 89 | 9274 | active |
| modelscope/modelscope ModelScope is a Python library and ecosystem built on the 'Model-as-a-Service' concept, providing unified APIs to download, run inference o… | 96 | 9149 | active |
| EvoMap/evolver Evolver is a Node.js-based self-evolving engine for AI agents built on the GEP (Gene Evolution Protocol), letting agents turn experience in… | 77 | 9145 | active |
| Nutlope/aicommits Aicommits is a Node.js CLI that automatically generates git commit messages from your staged changes using AI. It supports multiple LLM pro… | 94 | 9100 | active |
| FareedKhan-dev/kimi-k3-in-c A dependency-free C99 inference engine that runs the 2.78-trillion-parameter Kimi K3 model on a single CPU with as little as 8 GB of RAM by… | 79 | 9033 | active |
| deepseek-ai/DeepGEMM DeepGEMM is a high-performance CUDA BLAS kernel library for NVIDIA tensor cores, providing FP8, FP4, and BF16 GEMMs plus fused MoE and othe… | 82 | 8937 | active |
| Anxcye/anx-reader Anx Reader is a cross-platform e-book reader app built with Flutter that supports EPUB, MOBI, AZW3, FB2, TXT, and PDF formats. It integrate… | 81 | 8922 | active |
| Rig Rig is a Rust library for building modular, scalable LLM-powered applications and agents, offering a unified API across 20+ model providers… | 87 | 8911 | active |
| iflytek/astron-agent Astron Agent is an enterprise-grade, commercial-friendly agentic workflow platform for building, orchestrating, and deploying multi-agent A… | 83 | 8910 | active |
| BentoML BentoML is a Python framework for building online serving systems for AI apps and model inference, turning model inference scripts into RES… | 91 | 8909 | stable |
| Strands Agents Strands Agents is an open-source SDK (Python and TypeScript) for building production AI agents with a model-driven approach, supporting any… | 83 | 8884 | active |
| elder-plinius/OBLITERATUS OBLITERATUS is a Python toolkit for abliteration — identifying and removing refusal behaviors from large language models by locating and in… | 79 | 8861 | active |
| Nutlope/logocreator An open-source AI logo generator web app that creates brand-ready logos using FLUX models on Together AI, with logo editing, style presets,… | 62 | 8771 | active |
| freecodexyz/free-code A community fork of Anthropic's Claude Code terminal AI coding agent, rebuilt from an exposed source snapshot with telemetry removed, secur… | 54 | 8757 | active |
| a-ghorbani/pocketpal-ai PocketPal AI is an open-source mobile app that runs GGUF language models entirely on-device for private, offline chat. It supports model do… | 87 | 8727 | active |
| google-deepmind/alphafold3 Google DeepMind's official implementation of the AlphaFold 3 inference pipeline for predicting biomolecular structures and interactions. It… | 81 | 8643 | active |
| HKUDS/ClawWork ClawWork is an open-source benchmark and framework that turns AI assistants into 'AI coworkers' which must earn income by completing real p… | 44 | 8560 | active |
| OpenBMB/XAgent XAgent is an open-source, LLM-driven autonomous agent framework for automatically solving complex tasks. It features a dispatcher and plann… | 58 | 8551 | active |
| reorproject/reor Reor is a local-first, AI-powered desktop note-taking app that automatically links related notes, answers questions over your notes via RAG… | 10 | 8549 | active |
| bitsandbytes-foundation/bitsandbytes bitsandbytes is a Python library providing k-bit quantization primitives for PyTorch, enabling 8-bit (LLM.int8()) and 4-bit (QLoRA) quantiz… | 94 | 8514 | active |
| OptimalScale/LMFlow LMFlow is an extensible Python toolkit for finetuning and inference of large foundation models such as LLaMA, GPT-2, and Galactica. It prov… | 61 | 8487 | active |
| OGX OGX (formerly Llama Stack) is an open-source, OpenAI-compatible agentic API server and Python library that bundles inference, vector stores… | 87 | 8441 | active |
| qualcomm/GenieX GenieX is an on-device generative AI inference runtime for Qualcomm Snapdragon devices, running frontier LLMs and VLMs on the Hexagon NPU, … | 87 | 8440 | active |
| AAswordman/Operit Operit AI is an open-source AI agent and chat application for Android built with Kotlin and Jetpack Compose. It connects cloud or local LLM… | 86 | 8416 | active |
| withastro/flue Flue is an open-source TypeScript framework for building autonomous AI agents using a React-like hooks API ('use agent' directive, useModel… | 77 | 8415 | active |
| Dynamo NVIDIA Dynamo is an open-source, datacenter-scale distributed inference serving framework that orchestrates inference engines like vLLM, SG… | 88 | 8345 | active |
| zai-org/ChatGLM-6B ChatGLM-6B is an open-source bilingual (Chinese-English) 6.2B-parameter dialogue language model based on the GLM architecture, optimized fo… | 31 | 40938 | maintenance |
| anthropics/claude-agent-sdk-python The official Python SDK for building agents on top of Claude Code, providing async query APIs, bidirectional conversations, custom tools vi… | 83 | 8311 | active |
| duixcom/Duix-Mobile Duix Mobile is an open-source SDK for building real-time interactive AI avatars (digital humans) that run on-device on Android, iOS, tablet… | 63 | 8267 | active |
| k8sgpt k8sgpt is a CLI tool that scans Kubernetes clusters, diagnoses issues, and triages them in plain English using AI backends like OpenAI, Azu… | 98 | 8255 | active |
| mini-swe-agent A radically minimal AI software engineering agent (~100 lines of core Python) that solves GitHub issues and assists in the command line, sc… | 77 | 8226 | active |
| mini-swe-agent SWE-agent is a Python library that lets a large language model autonomously fix GitHub issues, discover security vulnerabilities, or run cu… | 77 | 8226 | active |
| intitni/CopilotForXcode An Xcode Source Editor Extension that brings GitHub Copilot, Codeium, and ChatGPT-powered AI assistance to Xcode on macOS. It provides inli… | 73 | 8208 | active |
| OpenDCAI/DataFlow DataFlow is a Python framework for preparing, cleaning, and synthesizing data for LLM training and RAG using composable LLM-based operators… | 77 | 8201 | active |
| AI4Finance-Foundation/FinRobot FinRobot is an open-source AI agent platform for financial applications built on large language models, unifying LLMs, reinforcement learni… | 78 | 8140 | active |
| NVIDIA/Isaac-GR00T NVIDIA Isaac GR00T N1.7 is an open vision-language-action (VLA) foundation model for generalized humanoid robot skills, taking language and… | 70 | 8116 | stable |
| InternLM/lmdeploy LMDeploy is a toolkit for compressing, quantizing, deploying, and serving large language models, built around its high-performance TurboMin… | 96 | 8107 | active |
| lm-sys/FastChat FastChat is an open platform for training, serving, and evaluating large language model based chatbots, originally releasing the Vicuna mod… | 54 | 39548 | maintenance |
| mattpocock/sandcastle A TypeScript library for orchestrating AI coding agents (like Claude Code) inside isolated sandboxes via a single sandcastle.run() call. It… | 69 | 8073 | active |
| Osaurus Osaurus is a native Swift macOS application that acts as a local-first AI harness, running open models on Apple Silicon (via MLX and Apple … | 82 | 8060 | active |
| PriorLabs/TabPFN TabPFN is a pre-trained tabular foundation model (transformer) that performs classification, regression, forecasting, and anomaly detection… | 96 | 8048 | active |
| RayVentura/ShortGPT ShortGPT is a Python framework that automates AI-driven video content creation, combining LLM script generation, voiceover synthesis, asset… | 31 | 7973 | active |
| OpenCoworkAI/open-codesign Open CoDesign is an open-source, local-first desktop AI design tool that turns prompts into polished artifacts like UI prototypes, slide de… | 79 | 7967 | active |
| Upsonic/Upsonic Upsonic is a Python framework for building autonomous and traditional AI agents, with support for tools, memory, knowledge bases, RAG, mult… | 71 | 7956 | active |
| ArvinLovegood/go-stock A Wails + NaiveUI desktop application that combines stock market data (A-shares, Hong Kong, US markets) with LLM-powered analysis, includin… | 85 | 7891 | active |
| chenyme/grok2api A self-hosted multi-account API gateway for Grok Build, Grok Web, and Grok Console, written in Go with a React frontend. It exposes Grok ca… | 82 | 7797 | active |
| 2FastLabs/agent-squad Agent Squad is an open-source framework for orchestrating multiple AI agents, routing each user query to the most suitable specialized agen… | 89 | 7780 | active |
| EricLBuehler/mistral.rs mistral.rs is a fast and flexible LLM inference engine written in Rust, supporting many model families with quantization (ISQ/UQFF, GGUF), … | 92 | 7741 | active |
| leejet/stable-diffusion.cpp A pure C/C++ inference engine for diffusion models (Stable Diffusion, FLUX, Wan, Qwen Image, Z-Image, and more) built on ggml in the style … | 92 | 7734 | active |
| sweepai/sweep Sweep is an AI coding assistant plugin for JetBrains IDEs (IntelliJ, PyCharm, GoLand, etc.) offering an AI agent, fast next-edit autocomple… | 36 | 7708 | active |
| farzaa/clicky Clicky is an open-source macOS app that acts as an AI teacher living next to your cursor — it can see your screen, talk with you via voice,… | 47 | 7707 | active |
| meta-llama/llama-models Meta's official repository of Llama large language models with Python utilities for working with them, including model cards, licenses, and… | 53 | 7693 | active |
| mufeedvh/code2prompt Code2Prompt is a Rust-based CLI tool that converts an entire codebase into a single structured LLM prompt, including a source tree, file co… | 66 | 7688 | active |
| tailcallhq/forgecode ForgeCode is an open-source, Rust-based CLI coding agent (AI pair programmer) that integrates LLMs from 300+ providers directly into the te… | 85 | 7637 | active |
| openJiuwen-ai/jiuwenswarm JiuwenSwarm is an open-source multi-agent AI system built on openJiuwen that lets a Leader agent decompose complex tasks and coordinate tea… | 82 | 7602 | active |
| pickle-com/glass Glass by Pickle is an open-source Electron desktop app that acts as a 'digital mind extension', seeing your screen and listening in real ti… | 39 | 7602 | active |
| lllyasviel/Omost Omost is a Python application that converts LLM coding capability into image composition by having pretrained LLMs (based on Llama3 and Phi… | 25 | 7595 | active |
| GoogleCloudPlatform/kubectl-ai kubectl-ai is an AI-powered Kubernetes assistant that translates natural-language user intent into precise kubectl operations. It is a Go C… | 71 | 7587 | active |
| lingfengQAQ/webnovel-writer A Claude Code plugin for long-form serialized webnovel writing, built in Python. It provides slash-command skills for initializing story se… | 84 | 7566 | active |
| mnfst/manifest Manifest is an open-source LLM gateway and router that lets AI agents and apps reach all their model providers through one OpenAI- and Anth… | 93 | 7551 | active |
| di-sukharev/opencommit OpenCommit is a CLI tool that uses LLMs (OpenAI, Claude, Ollama, and other providers) to automatically generate meaningful git commit messa… | 66 | 7545 | active |
| Refly Refly is an open-source platform for building, versioning, and running reusable AI agent skills through a natural-language 'vibe workflow' … | 74 | 7531 | active |
| guardrails-ai/guardrails Guardrails is a Python framework for building reliable LLM applications by running Input/Output Guards that detect and mitigate risks, and … | 95 | 7527 | active |
| teamchong/pxpipe pxpipe is a local TypeScript proxy that reduces Claude Code input token usage by rewriting bulky text context (system prompts, tool docs, h… | 81 | 7525 | active |
| algorithmicsuperintelligence/openevolve OpenEvolve is an open-source Python implementation of DeepMind's AlphaEvolve, an evolutionary coding agent that uses LLM ensembles to auton… | 86 | 7518 | active |
| microsoft/graphrag GraphRAG is a Python library and CLI pipeline from Microsoft Research that extracts a knowledge graph from unstructured text, builds commun… | 92 | 36292 | maintenance |
| zai-org/GLM-OCR GLM-OCR is an open-source 0.9B-parameter multimodal OCR model built on the GLM-V encoder-decoder architecture for complex document understa… | 73 | 7504 | active |
| open-compass/opencompass OpenCompass is a Python-based evaluation platform and toolkit for benchmarking large language models across 100+ datasets, supporting model… | 93 | 7471 | active |
| apple/ml-fastvlm Official implementation of FastVLM, a vision language model with an efficient hybrid vision encoder (FastViTHD) that reduces token count an… | 62 | 7423 | active |
| MuiseDestiny/zotero-gpt A Zotero plugin that integrates OpenAI GPT models into the Zotero reference manager, letting users ask questions about PDFs, summarize pape… | 73 | 7420 | active |
| spinabot/brigade Brigade is a self-hosted, MIT-licensed multi-agent runtime where a crew of AI agents on an org chart delegate work, switch models mid-task,… | 73 | 7412 | active |
| vladmandic/sdnext SD.Next is an open-source, self-hosted WebUI server application for AI generative image and video creation built on Stable Diffusion and Di… | 77 | 7353 | active |
| zai-org/GLM-5 GLM-5 is a family of open-weight large language models from Z.ai (Zhipu AI), including the flagship GLM-5.2 with 1M-token context and the m… | 60 | 7329 | active |
| interviewstreet/hiring-agent An open-source AI agent pipeline from HackerRank that extracts structured data from resume PDFs, enriches it with GitHub signals, and produ… | 63 | 7312 | active |
| kyegomez/swarms Swarms is an enterprise-grade Python framework for orchestrating multi-agent systems powered by large language models, offering prebuilt se… | 66 | 7282 | active |
| Paddle Lite Paddle Lite is a high-performance, lightweight deep learning inference engine from Baidu's PaddlePaddle ecosystem, designed for mobile, emb… | 55 | 7277 | active |
| JerryZLiu/Dayflow Dayflow is an open-source, local-first macOS app that automatically turns your screen activity into a timeline of what you accomplished, us… | 81 | 7276 | active |
| InternLM/InternLM Official repository for the InternLM series of open-source large language models (InternLM, InternLM2, InternLM2.5, InternLM3), including p… | 38 | 7276 | active |
| mit-han-lab/streaming-llm StreamingLLM is a research framework from MIT Han Lab implementing the Attention Sinks method (ICLR 2024) for efficient streaming language … | 28 | 7267 | stable |
| ml-explore/mlx-lm A Python package for running and fine-tuning large language models on Apple silicon using the MLX framework. It integrates with the Hugging… | 78 | 7259 | active |
| magnitudedev/magnitude Magnitude is an open-source CLI tool and agent that bundles local LLM inference with an agent harness, profiling your hardware to recommend… | 82 | 7240 | active |
| MineDojo/Voyager Voyager is an open-source LLM-powered embodied agent that plays Minecraft through GPT-4, using an automatic curriculum, an ever-growing ski… | 30 | 7219 | active |
| deepseek-ai/DeepSpec DeepSpec is a full-stack Python codebase from DeepSeek for training and evaluating draft models used in speculative decoding of large langu… | 51 | 7207 | active |
| NVIDIA-NeMo/Guardrails NVIDIA NeMo Guardrails is an open-source Python toolkit for adding programmable guardrails to LLM-based conversational applications. It sit… | 90 | 7181 | active |
| Automaton Automaton is a TypeScript runtime for continuously running, self-improving AI agents that pay for their own compute via an Ethereum wallet … | 67 | 7156 | active |
| dwzhu-pku/PaperBanana PaperBanana is a reference-driven multi-agent framework that automatically generates publication-ready academic illustrations such as metho… | 52 | 7150 | active |