Local LLM projects on GitHub

1. 1
   #1 · openclaw/openclaw

   openclaw

   An open-source AI assistant that runs on your own computer and meets you in the chat apps you already use — WhatsApp, Telegram, Discord, iMessage, Slack and 20+ more.

   392.2K
2. 2
   #2 · ollama/ollama

   ollama

   Ollama is a tool for downloading and running open large language models like DeepSeek, Qwen, Gemma, Llama, Mistral and gpt-oss on your own machine, with a simple CLI and a local REST API.

   183.5K
3. 3
   #3 · ggml-org/llama.cpp

   llama.cpp

   llama.cpp is a plain C/C++ implementation for running LLM (and VLM) inference with minimal setup and state-of-the-art performance, either locally or in the cloud.

   131.4K

4. 4

   TradingAgents

   111.6KTauricResearch/TradingAgents · Python

   TradingAgents is a multi-agent LLM framework that simulates a trading firm: specialized agents analyze fundamentals, sentiment, news, and technicals, bull/bear researchers debate the findings, and a trader plus risk/portfolio manager team decide whether to execute the trade.

   • multiagent
   • backtesting
   • algorithmic-trading
   • stock-market
5. 5

   odysseus

   93.7Kodysseus-dev/odysseus · Python

   Odysseus is a self-hosted AI workspace that bundles chat, agents, deep research, a document editor, email, notes, tasks, and calendar around local or API models.

   • imap
   • smtp
   • web-search
   • calendar
6. 6

   NextChat

   88.8KChatGPTNextWeb/NextChat · TypeScript

   NextChat is a fast, lightweight AI chat assistant that lets you talk to GPT-4o, Claude, Gemini, DeepSeek and 100+ other models through one unified interface, deployable with a single click or self-hosted.

   • gemini-pro
   • gpt-4o
   • groq
   • tauri-app
7. 7

   worldmonitor

   88.6Kkoala73/worldmonitor · TypeScript

   World Monitor is a real-time global intelligence dashboard that aggregates and AI-synthesizes news, geopolitical signals, and infrastructure data into one situational-awareness interface.

   • news-aggregator
   • news
   • opensource
   • openapi
8. 8

   unsloth

   78.2Kunslothai/unsloth · Python

   Unsloth is a local-first app for running and training LLMs, diffusion, TTS and embedding models on your own machine, including models like Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4 and FLUX.

   • dpo
   • qlora
   • intel
   • kaggle
9. 9

   gpt4all

   77.4Knomic-ai/gpt4all · C++

   GPT4All is a desktop application that runs large language models privately on everyday laptops and desktops, with no API calls or GPU required.

   • ai-chat
   • llama-cpp
   • gguf
   • langchain
10. 10

    ruflo

    75.8Kruvnet/ruflo · TypeScript

    Ruflo is an agent meta-harness for Claude Code and Codex: it adds 100+ specialized agents, coordinated swarms, self-learning memory, and cross-machine federation on top of those coding agents so they can collaborate instead of running solo.

    • swarm-intelligence
    • swarm
    • agentic-framework
    • federation
11. 11

    anything-llm

    67.2KMintplex-Labs/anything-llm · JavaScript

    AnythingLLM is an all-in-one, self-hosted AI application that lets you chat with your documents, run AI agents, and build a private ChatGPT-like experience using local or cloud LLMs.

    • localai
    • self-hosted-ai
    • agentic-workflow
    • hermes-agent
12. 12

    free-claude-code

    61KAlishahryar1/free-claude-code · Python

    FCC (Free Claude Code) routes coding agents like Claude Code, Codex, and OpenCode to 50 ToS-friendly providers, giving you 1.3B+ free tokens every month.

    • failover
    • multi-model
    • jetbrains
    • llm-gateway
13. 13

    serviette

    58.8Kpathwaycom/serviette · Python

    This repo provides ready-to-run cloud templates for RAG pipelines, AI pipelines, and enterprise search, built on the Pathway Live Data Framework and kept in sync with live data sources.

    • enterprise-search
    • pinecone
    • milvus
    • chromadb
14. 14

    voicebox

    57.8Kjamiepine/voicebox · Python

    Voicebox is an open-source, local-first AI voice studio: it clones voices, generates speech in 23 languages, lets you dictate into any app with a global hotkey, and gives MCP-aware AI agents a voice - all running on your own machine as a free alternative to ElevenLabs and WisprFlow combined.

    • voice-clone
    • voice-ai
    • mlx
    • whisper
15. 15

    private-gpt

    57.6Kzylon-ai/private-gpt · Python

    PrivateGPT is an open-source API layer that turns any local, OpenAI-compatible model into a full production-ready AI application backend, covering RAG, tools, MCP, text-to-sql and more.

    • on-premise
    • text-to-sql
    • agentic-rag
    • ai-tools
16. 16

    LocalAI

    49.6Kmudler/LocalAI · Go

    LocalAI is an open-source, self-hosted AI engine that lets you run LLMs, vision, voice, image and video models on your own hardware, including CPU-only setups, through an OpenAI/Anthropic/ElevenLabs-compatible API.

    • rerank
    • text-generation
    • face-recognition
    • distributed
17. 17

    nanobot

    49.1KHKUDS/nanobot · Python

    nanobot is an ultra-lightweight, open-source, self-hosted personal AI agent framework written in Python that runs as a browser WebUI, terminal client, or chat-app integration, combining tools, long-term memory, MCP, multi-agent delegation, and scheduled automation behind an OpenAI-compatible API.

    • personal-ai-assistant
    • discord-bot
    • feishu
    • llms
18. 18

    exo

    47.9Kexo-explore/exo · Python

    exo turns a group of your own devices (Macs, Linux boxes) into a single AI cluster so you can run large AI models that would never fit on one machine, pooling their memory and compute together.

    • kimi
    • distributed
    • mlx
    • model-serving
19. 19

    textgen

    47.7Koobabooga/textgen · Python

    TextGen is an open-source desktop app for running local LLMs with text, vision, tool-calling, and an OpenAI/Anthropic-compatible API, built to be 100% private with no telemetry.

    • vision
    • rocm
    • llama-cpp
    • tool-calling
20. 20

    jan

    45Kjanhq/jan · Rust

    Jan is an open-source, self-hosted alternative to ChatGPT that runs 100% offline on your own computer, letting you download and run LLMs locally with full control and privacy.

    • localai
    • llamacpp
    • gguf
    • tauri
21. 21

    Fabric

    44.3Kdanielmiessler/Fabric · Go

    Fabric is an open-source framework that organizes AI prompts (called Patterns) by real-world task, so you can apply AI to specific problems from the command line or via a REST API.

    • augmentation
    • azure
    • raspberry-pi
    • ollama
22. 22

    colibri

    42.5KJustVugg/colibri · C

    Colibri is an open-source engine for running frontier mixture-of-experts (MoE) language models on hardware you already own — written in pure C with zero dependencies.

    • on-device-ai
    • on-device
    • inference-engine
    • mixture-of-experts
23. 23

    chatbox

    42Kchatboxai/chatbox · TypeScript

    Chatbox is a desktop AI client that lets you chat with ChatGPT, Claude, Gemini, local Ollama models and other LLMs from one unified app, available on Windows, Mac, Linux, iOS, Android and the web.

    • gpt-5
    • assistant
    • copilot
    • chat
24. 24

    BitNet

    41.2Kmicrosoft/BitNet · C++

    bitnet.cpp is Microsoft's official inference framework for 1-bit (ternary, 1.58-bit) large language models like BitNet b1.58, letting them run fast on CPUs and GPUs.

    • cpu
    • x86
    • arm
    • on-device
25. 25

    project-nomad

    40.4KCrosstalk-Solutions/project-nomad · TypeScript

    Project NOMAD is a self-contained, offline-first knowledge and education server: offline Wikipedia, thousands of books, Khan Academy courses, offline maps and optional local AI, all running on hardware you own with no internet required.

    • maps
    • homelab
    • offline-first
    • openai-compatible
26. 26

    OpenViking

    40.1Kvolcengine/OpenViking · Python

    An open-source context database for AI agents: memories, resources, and skills live in one virtual filesystem under the viking:// protocol, so the agent browses its own context with ls, tree, and find instead of querying a black-box vector store.

    • agentic-rag
    • token-optimization
    • self-evolving
    • memory-management
27. 27

    quivr

    39.6KThe-Vibe-Company/quivr · Go

    Quivr is an open-source engine that turns your files and content streams into search and monitoring — a 'second brain' you can query, built around an opinionated RAG powered by generative AI.

    • groq
    • vector
    • data-pipeline
    • hybrid-search
28. 28

    airllm

    39.1Klyogavin/airllm · Jupyter Notebook

    An open-source Python library that runs huge LLMs on tiny GPUs: a 70B model fits on a single 4GB card without quantization, distillation or pruning — even the 2.8T-parameter Kimi K3 runs in under 4GB.

    • consumer-gpu
    • chinese-nlp
    • qlora
    • mlx
29. 29

    llmfit

    38KAlexsJones/llmfit · Rust

    llmfit inspects your CPU, RAM, GPU(s), VRAM, and accelerator setup, then tells you which open-source LLMs and quantizations will actually run well on your machine.

    • localai
    • llama-cpp
    • mlx
    • gguf
30. 30

    khoj

    37.7Kkhoj-ai/khoj · Python

    Khoj is an open-source personal AI — a 'second brain' that answers your questions from the web and your own documents, runs custom agents for you, and works with any online or local LLM.

    • obsidian-md
    • personal-ai-assistant
    • stt
    • llama3
31. 31

    Vane

    37.4KItzCrazyKns/Vane · TypeScript

    Vane is an open-source, privacy-focused AI answering engine: it searches the web for you and answers with cited sources, running entirely on your own hardware.

    • self-hosted-ai
    • search-engine
    • openai-compatible
    • ollama
32. 32

    heretic

    35.6Kp-e-w/heretic · Python

    Heretic is a command-line tool that automatically removes censorship (safety alignment) from transformer-based language models using directional ablation combined with an Optuna-powered parameter optimizer, without any training.

    • uncensored
    • optimization
    • evaluation
    • quantization
33. 33

    openclaude

    34KTwigpine/openclaude · TypeScript

    OpenClaude is an open-source, terminal-first coding-agent CLI that can connect to many cloud and local model providers (OpenAI-compatible APIs, Gemini, GitHub Models, Codex, Ollama, and more) through a single consistent workflow of prompts, tools, agents, MCP, and slash commands.

    • coding
    • vscode-extension
    • ai-tools
    • ollama
34. 34

    tabby

    33.9KTabbyML/tabby · Rust

    Tabby is a self-hosted AI coding assistant that gives you an open-source, on-premises alternative to GitHub Copilot, with code completion and chat features.

    • code-completion
    • codegen
    • developer-experience
    • intellij
35. 35

    AionUi

    33.6KiOfficeAI/AionUi · TypeScript

    AionUi is a free, open-source Cowork app for AI agents: they work alongside you on your computer — reading files, writing code, browsing the web, automating tasks 24/7 — while you see everything they do and stay in control.

    • agent-team
    • nano-banana
    • clawdbot
    • feishu
36. 36

    chatbot-ui

    33.3Kmckaywrigley/chatbot-ui · TypeScript

    An open-source AI chat web app — a ChatGPT-style interface you can self-host and point at different AI models.

    • supabase
    • postgres
    • ollama
    • nextjs
37. 37

    self-llm

    32.5Kdatawhalechina/self-llm · Jupyter Notebook

    Self-llm is a Chinese-language tutorial collection that teaches beginners how to configure environments, deploy, and fine-tune (full-parameter or LoRA) 50+ open-source large language models and multimodal models on Linux.

    • deepseek-r1
    • lmstudio
    • chatglm
    • llama3
38. 38

    meetily

    31.8KZackriya-Solutions/meetily · Rust

    A privacy-first AI meeting assistant that captures your meetings, transcribes them in real time, and generates summaries — entirely on your own machine, with no data sent to the cloud.

    • diarization
    • meeting-notes
    • asr
    • on-device
39. 39

    colima

    31.2Kabiosoft/colima · Go

    Colima provides container runtimes on macOS (and Linux) with minimal setup: it starts a Linux VM running Docker, Containerd or Incus, all managed through one simple command.

    • containerd
    • k8s
    • runtime
    • virtualization
40. 40

    qmd

    30.5Ktobi/qmd · TypeScript

    QMD is an on-device search engine for your personal knowledge: it indexes markdown notes, meeting transcripts, documentation and knowledge bases, and searches them with keywords or natural language.

    • rerank
    • retrieval
    • hybrid-search
    • bm25
41. 41

    deepagents

    30.3Klangchain-ai/deepagents · Python

    Deep Agents is an open source, batteries-included agent harness from LangChain: an opinionated agent that runs out of the box with sub-agents, a filesystem, context management, and skills bundled in — and every piece can be extended, overridden, or replaced without forking.

    • deepagents
    • langgraph
    • mcp-client
    • harness-engineering
42. 42

    picoclaw

    30.1Ksipeed/picoclaw · Go

    An open-source personal AI assistant written entirely in Go by Sipeed: connect it to an LLM provider and chat with it from Telegram, Discord, Matrix, WeChat or the built-in Web UI.

    • risc-v
    • assistant
    • cron
    • webui
43. 43

    sim

    29.8Ksimstudioai/sim · TypeScript

    Sim is a collaborative workspace for building, deploying, and monitoring AI agents and workflows, used by 100,000+ builders. Agents can be built visually, in chat, or with code.

    • aiagents
    • agentic-workflow
    • low-code
    • no-code
44. 44

    karakeep

    29.7Kkarakeep-app/karakeep · TypeScript

    Karakeep (formerly Hoarder) is a self-hostable bookmark-everything app: save links, notes, images and PDFs, and it auto-fetches link details, tags everything with an LLM and gives you full-text and semantic search over all of it.

    • read-it-later
    • web-archiving
    • bookmarks
    • yt-dlp
45. 45

    void

    28.8Kvoideditor/void · TypeScript

    Void was an open-source AI code editor forked from VS Code: it runs AI agents on your codebase, checkpoints and visualizes changes, and lets you bring any model or host one locally.

    • visual-studio-code
    • copilot
    • vscode-extension
    • ide
46. 46

    Resume-Matcher

    28.7Ksrbhr/Resume-Matcher · Python

    Resume Matcher is a free, open-source AI harness for tailoring your resume to each job application, working with Claude, ChatGPT, DeepSeek, Kimi, GLM, Gemma and other LLMs, both local and remote.

    • resume-builder
    • word-embeddings
    • resume
    • job-search
47. 47

    crush

    28.7Kcharmbracelet/crush · Go

    A terminal-based AI coding agent from Charm: it wires your tools, code, and workflows into the LLM of your choice and works as a coding partner right in your terminal.

    • lsp
    • agentic-coding
    • hooks
    • llms
48. 48

    agenticSeek

    27.5KFosowl/agenticSeek · Python

    AgenticSeek is a fully local, voice-enabled alternative to Manus AI: an autonomous agent that browses the web, writes code and plans tasks while keeping all data on your own hardware.

    • deepseek-r1
    • voice-assistant
    • local-ai
    • llm-agents
49. 49

    haystack

    26.8Kdeepset-ai/haystack · Python

    An open-source AI orchestration framework for building production-ready LLM applications in Python: modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation.

    • agentic-rag
    • information-retrieval
    • genai
    • retrieval-augmented-generation
50. 50

    Strata

    26.4KNiko1221/Strata · C++

    Strata runs Qwen3.8-Flash-Next, a 125-billion-parameter AI model, on a normal gaming PC with one NVIDIA GPU (12-24GB VRAM) and 64GB of RAM, with a one-click installer for Windows and Linux.

    • llama-cpp
    • speculative-decoding
    • inference-engine
    • mixture-of-experts
51. 51

    llamafile

    26.4Kmozilla-ai/llamafile · C++

    llamafile turns open LLMs into single-file executables: one file that runs a chat model locally on most operating systems and CPU architectures, with no installation. It is a Mozilla Builders project, now revamped by Mozilla.ai.

    • llamacpp
    • llama-cpp
    • gguf
    • local-ai
52. 52

    kotaemon

    25.8KCinnamon/kotaemon · Python

    Kotaemon is an open-source RAG tool for chatting with your documents: a clean web UI where you ask questions about your own files, serving both end users and developers.

    • document-parsing
    • question-answering
    • llamacpp
    • retrieval-augmented-generation
53. 53

    ds4

    23.9Kantirez/ds4 · C

    DwarfStar (ds4) is a small, self-contained local inference engine specialized for running DeepSeek V4 Flash and PRO, plus GLM 5.2/5.3, on consumer hardware like high-memory Macs, NVIDIA GPUs, and AMD Strix Halo systems.

    • kv-cache
    • llm-serving
    • amd-gpu
    • glm
54. 54

    omlx

    22.9Kjundot/omlx · Python

    oMLX is a local LLM inference server for Apple Silicon Macs. It serves text LLMs, vision models, OCR, embeddings and rerankers with continuous batching and a tiered KV cache, managed from the macOS menu bar.

    • menubar-app
    • vlm
    • mlx
    • model-serving
55. 55

    Awesome-Chinese-LLM

    22.8KAiHubCN/Awesome-Chinese-LLM

    A curated collection of open-source LLM resources for Chinese, focused on smaller models that can be privately deployed and trained cheaply. It lists 100+ resources: models, applications, datasets, and tutorials.

    • chatglm
    • minimax
    • glm
    • awesome-lists
56. 56

    waveterm

    22.5Kwavetermdev/waveterm · Go

    Wave is an open-source, AI-integrated terminal for macOS, Linux, and Windows. It bundles a context-aware AI assistant, durable SSH sessions, and a block-based workspace with editors and file previews.

    • secret-management
    • file-management
    • terminal-emulators
    • remote-access
57. 57

    localGPT

    22.2KPromtEngineer/localGPT · Python

    LocalGPT is a fully private, on-premise document intelligence platform: chat with, summarise and query your own documents using LLMs that run locally, so no data ever leaves your machine.

    • document-intelligence
    • document
    • full-text-search
    • retrieval-augmented-generation
58. 58

    guidance

    21.8Kguidance-ai/guidance · Jupyter Notebook

    Guidance is a Python library — an efficient programming paradigm for steering large language models: you control how output is structured while reducing latency and cost compared with conventional prompting or fine-tuning.

    • large-language-model
    • grammar
    • regex
    • llama-cpp
59. 59

    candle

    21.2Khuggingface/candle · Rust

    Candle is a minimalist machine learning framework for Rust, focused on performance (including GPU support) and ease of use. It ships with ready examples for LLMs, speech, vision and text-to-image models.

    • yolo
    • stable-diffusion
    • gguf
    • whisper
60. 60

    gpt-oss

    20.5Kopenai/gpt-oss · Python

    The gpt-oss series: two open-weight language models from OpenAI — gpt-oss-120b and gpt-oss-20b — built for powerful reasoning, agentic tasks, and versatile developer use cases. The repo itself adds reference implementations, tools, and example clients around the models.

    • chain-of-thought
    • lmstudio
    • gpt-oss
    • function-calling
61. 61

    DB-GPT

    20.2Keosphoros-ai/DB-GPT · Python

    An open-source agentic AI data assistant: connect databases, files or knowledge bases, ask questions in plain language, and the AI writes SQL and code, runs the analysis, and produces charts, dashboards and reports.

    • chromadb
    • text-to-sql
    • multi-model
    • gpt-4
62. 62

    eliza

    19.6KelizaOS/eliza · TypeScript

    elizaOS is an open-source TypeScript framework and product stack for autonomous AI agents: the core runtime, the Eliza app, the CLI, cloud services, native bridges, and first-party plugins in one monorepo. Bootable OS builds live in a separate elizaOS/os repo.

    • swarm
    • solana
    • wallet
    • crypto
63. 63

    pyvideotrans

    19.4Kjianchang512/pyvideotrans · Python

    An open-source video translation and dubbing tool: it converts a video from one language to another through a one-click pipeline of speech recognition, subtitle translation, TTS dubbing, and video synthesis.

    • dubbing
    • srt
    • stt
    • speaker-diarization
64. 64

    web-llm

    19.3Kmlc-ai/web-llm · TypeScript

    WebLLM runs large language models directly in the browser, accelerated by WebGPU — no server involved. It exposes the OpenAI API, so standard OpenAI-style code works against open models running locally.

    • json-schema
    • on-device-ai
    • webgpu
    • inference-engine
65. 65

    kirara-ai

    19.1Klss233/kirara-ai · Python

    Kirara AI is a DIY, open-source multimodal AI chatbot you self-host: a Python server that plugs mainstream LLMs into chat platforms like QQ, Telegram, Discord and WeChat.

    • chatbot-framework
    • qq
    • stable-diffusion
    • bot
66. 66

    Chinese-LLaMA-Alpaca

    18.9Kymcui/Chinese-LLaMA-Alpaca · Python

    Open-source Chinese LLaMA and Alpaca large language models. The base LLaMA was extended with a Chinese vocabulary and further pre-trained on Chinese data, and Alpaca versions were instruction-tuned on Chinese instructions; training and deployment scripts are open too.

    • llama3
    • llama2
    • instruction-tuning
    • pretraining
67. 67

    WeClone

    18.3Kxming521/WeClone · Python

    WeClone is an open-source toolkit that turns your chat history into an AI twin: it fine-tunes an LLM on your own messages so a chatbot can reply in your personal style.

    • ai-companion
    • lora
    • slack
    • discord
68. 68

    LangBot

    18.2Klangbot-app/LangBot · Python

    An open-source, production-grade platform for building AI-powered instant messaging bots: it connects LLMs to chat platforms so you can ship agents that converse, execute tasks, and plug into your workflows.

    • coze
    • dify
    • qq
    • feishu
69. 69

    rowboat

    18Krowboatlabs/rowboat · TypeScript

    Rowboat is a desktop AI coworker with a memory of your work. It indexes your email, meetings and chats into a living knowledge graph and acts through built-in surfaces: email, notes, browser, code mode, meeting notes and workspaces.

    • meeting-notes
    • claude-cowork
    • multiagent
    • knowledge-management
70. 70

    leon

    17.6Kleon-ai/leon · TypeScript

    Leon is an open-source personal AI assistant that runs locally and is built around tools, context, memory, and agentic execution — it understands a goal, uses real tools, and completes tasks end to end.

    • personal-assistant
    • assistant
    • skill-md
    • voice-assistant
71. 71

    DeepCode

    16.7KHKUDS/DeepCode · Python

    An open agentic coding platform from HKUDS: a multi-agent harness that turns natural language and even research papers into production-ready code, working directly in your repository.

    • llm-agent
    • harness-engineering
    • agentic-coding
    • human-in-the-loop
72. 72

    web-ui

    16.6Kbrowser-use/web-ui · Python

    A Gradio web UI for the browser-use library: give a task to an AI agent and watch it drive a real browser. It supports many LLMs — OpenAI, Anthropic, Google, Azure OpenAI, DeepSeek, Ollama and more.

    • cloud-browser
    • screen-recording
    • vnc
    • gradio
73. 73

    pipecat

    16.5Kpipecat-ai/pipecat · Python

    Pipecat is an open-source Python framework for building real-time voice and multimodal conversational agents — from a single voice bot to a distributed multi-agent system — maintained by Daily and the community.

    • chatbot-framework
    • chatbots
    • multimodal-ai
    • ai-companion
74. 74

    SurfSense

    16.5KMODSetter/SurfSense · Python

    An open-source, self-hostable NotebookLM alternative: agents pull structured data from the live web (Reddit, YouTube, Amazon and more) through connectors, and a built-in knowledge base keeps every finding searchable with citations.

    • notebooklm
    • self-hosted-ai
    • full-text-search
    • langchain
75. 75

    OpenHarness

    16KHKUDS/OpenHarness · Python

    OpenHarness is an open-source Python agent harness: the infrastructure that turns an LLM into a working agent with tool use, skills, memory and multi-agent coordination. It also ships ohmo, a personal agent that works for you in chat apps.

    • context-management
    • subagents
    • hooks
    • agent-harness
76. 76

    Second-Me

    15.7Kmindverse/Second-Me · Python

    Second Me is an open-source project that lets you train a personal, locally-hosted AI model that preserves your identity, memory, and context, then connects it to a decentralized network so it can represent you across apps.

    • personal-ai
    • llamacpp
    • mlx
    • decentralized
77. 77

    RWKV-LM

    14.8KBlinkDL/RWKV-LM · Python

    RWKV (pronounced 'RwaKuv') is an RNN architecture with transformer-level LLM performance that can also be trained in parallel like GPT. The current generation, RWKV-7 'Goose', is 100% attention-free, linear-time and constant-memory (no kv-cache).

    • gpt-2
    • gpt-3
    • rnn
    • lstm
78. 78

    FreeToken

    14.5KFlashML-org/FreeToken · Python

    FreeToken is an edge-native Mixture-of-Experts (MoE) serving engine built to run frontier-scale open-weight models, including 290B+ parameter models, on consumer hardware like gaming PCs.

    • edge
    • minimax
    • glm
    • edge-ai
79. 79

    hermes-desktop

    14.4Kfathah/hermes-desktop · TypeScript

    Hermes One is a community-maintained native desktop app for installing, configuring, and chatting with Hermes Agent — a self-improving AI assistant with tool use, multi-platform messaging, and a closed learning loop.

    • hermes
    • cron
    • hermes-agent
    • slack
80. 80

    QAnything

    14.2Knetease-youdao/QAnything · Python

    A local knowledge base question-answering system: load your own documents into it and ask questions — you get accurate answers drawn from your files instead of the open web.

    • question-answering
    • vector-search
    • offline-first
    • semantic-search
81. 81

    MoneyPrinter

    14.1KFujiwaraChoki/MoneyPrinter · Python

    MoneyPrinter automates the creation of YouTube Shorts: you give it a topic and it generates the script, voice, and final video using MoviePy, with everything powered by local Ollama models.

    • youtube
    • content-creation
    • video-generation
    • chatgpt
82. 82

    needle

    14.1Kcactus-compute/needle · Python

    Needle 2 is an open 45M-parameter model for tool calling, device use and structured extraction that ships as a single 14MB binary and runs a full session in about 28MB of RAM. This repo is its Python package for inference, LoRA fine-tuning and export.

    • jax
    • function-calling
    • on-device-ai
    • tool-calling
83. 83

    dalai

    12.9Kcocktailpeanut/dalai · CSS

    Dalai is a tool that lets you download and run LLaMA or Alpaca language models directly on your own computer, with a web UI, a CLI, and a JavaScript/socket.io API, built on top of llama.cpp and alpaca.cpp.

    • socket-io
    • llama
    • webui
    • websocket
84. 84

    llm

    12.6Ksimonw/llm · Python

    LLM is a CLI tool and Python library for running prompts against OpenAI, Anthropic Claude, Google Gemini, Qwen, Gemma, Kimi, DeepSeek, Mistral and dozens of other large language models, either via remote APIs or models running locally on your own machine.

    • function-calling
    • llms
    • embeddings
    • ollama
85. 85

    OpenLLM

    12.6Kbentoml/OpenLLM · Python

    OpenLLM lets you run open-source LLMs like Llama, DeepSeek, Mistral and Qwen as OpenAI-compatible API endpoints with a single command, including a built-in chat UI and cloud deployment support.

    • llm-ops
    • llama2
    • mistral
    • llm-serving
86. 86

    keep

    12.4Kkeephq/keep · Python

    Keep is an open-source AIOps and alert management platform: a single pane of glass for alerts and incidents, with deduplication, enrichment, filtering, correlation, workflows and dashboards built in.

    • monitoring-tool
    • sre
    • alerting
    • incident-response
87. 87

    h2ogpt

    11.9Kh2oai/h2ogpt · Python

    h2oGPT is an open-source private chat application that lets you query and summarize your own documents or just chat with local LLMs, with no data leaving your machine.

    • generative
    • llama2
    • llamacpp
    • stable-diffusion
88. 88

    pal-mcp-server

    11.8KBeehiveInnovations/pal-mcp-server · Python

    PAL MCP (formerly Zen MCP) is a Model Context Protocol server that lets your AI coding CLI (Claude Code, Gemini CLI, Codex CLI, etc.) orchestrate multiple AI models (Gemini, OpenAI, Azure, Grok, OpenRouter, Ollama, local models) within a single conversation.

    • model-context-protocol
    • orchestration
    • agent-orchestration
    • ollama
89. 89

    MiniCPM

    11.6KOpenBMB/MiniCPM · Jupyter Notebook

    MiniCPM is OpenBMB's family of small, open-source LLMs built for on-device and resource-constrained use. The newest release, MiniCPM5-2B, is a dense 2B-parameter model that reaches open-source state-of-the-art for its size class while staying competitive with much larger 4B-class models.

    • tool-use
    • knowledge-distillation
    • long-context
    • llamacpp
90. 90

    brigade

    11.3Kspinabot/brigade · TypeScript

    Brigade is a self-hosted personal AI crew: a team of agents that share one long-term memory (Tideline), delegate tasks to each other on a real org chart, and act across 1,000+ apps you already use.

    • moltbot
    • agent-runtime
    • clawdbot
    • hermes
91. 91

    presenton

    11.2Kpresenton/presenton · TypeScript

    Presenton is an open-source AI presentation generator: it turns a prompt or an uploaded document into a fully editable PowerPoint (.pptx) or PDF deck. It is a self-hostable alternative to tools like Gamma and Canva.

    • presentation
    • byok
    • ollama
    • electron
92. 92

    hive

    11.1Kaden-hive/hive · Python

    OpenHive is an open-source multi-agent harness for production AI — a runtime that adds state management, failure recovery, observability, cost control, and human oversight so colonies of agents reliably run real business processes instead of demos.

    • self-improving
    • fault-tolerance
    • cost-tracking
    • harness-engineering
93. 93

    llama-gpt

    10.9Kgetumbrel/llama-gpt · TypeScript

    LlamaGPT is a self-hosted, offline, ChatGPT-like chatbot powered by Llama 2 (and now Code Llama), so you get a familiar chat UI without sending any data to a third-party cloud.

    • localai
    • llama2
    • gpt-4
    • llamacpp
94. 94

    VidBee

    10.9Knexmoe/VidBee · TypeScript

    A free, open-source desktop app that downloads video and audio from 1000+ sites or imports local files, transcribes them on your machine, and lets you summarize, translate, or ask questions about them with the AI provider you choose.

    • facebook
    • speaker-diarization
    • tiktok
    • yt-dlp
95. 95

    mistral-inference

    10.8Kmistralai/mistral-inference · Jupyter Notebook

    The official inference code for Mistral AI's open-weight models: minimal Python and CLI tooling to run models like Mixtral, Codestral, Pixtral and Mistral Large locally.

    • mistral
    • coding-assistant
    • function-calling
    • open-weights
96. 96

    ai-town

    10.7Ka16z-infra/ai-town · TypeScript

    AI Town is an MIT-licensed starter kit for building a virtual town where AI characters live, chat, and socialize with each other.

    • starter-kit
    • canvas
    • simulation
    • ollama
97. 97

    petals

    10.6Kbigscience-workshop/petals · Python

    Petals lets you run big open LLMs at home, BitTorrent-style: you join a public swarm that serves models like Llama 3.1 (405B), Mixtral (8x22B), Falcon (40B+) and BLOOM (176B) for inference and fine-tuning right from your desktop or Google Colab, up to 10x faster than offloading.

    • distributed-computing
    • colab
    • community-driven
    • open-weights
98. 98

    ollama-python

    10.6Kollama/ollama-python · Python

    This is the official Python library for Ollama, giving Python 3.8+ projects an easy way to call a local or cloud Ollama server to run large language models.

    • gpt-oss
    • http-client
    • async
    • llm-inference
99. 99

    aichat

    10.5Ksigoden/aichat · Rust

    AIChat is an all-in-one LLM command-line tool: a shell assistant, a chat REPL, RAG, AI tools and agents, with access to OpenAI, Claude, Gemini, Ollama, Groq and more.

    • repl
    • multi-provider
    • llm-proxy
    • mistral
100. 100

     inkos

     10.4KNarcooo/inkos · TypeScript

     InkOS is an open-source story-creation AI agent system: it plans, writes, reviews and revises long serialized novels, short stories, screenplays, storyboards, interactive film-games and translations, all from one workbench.

     • chinese-novel
     • lmstudio
     • storyboard
     • short-drama

Showing the 100 most-starred of 218. See every one in the catalog.

• llm1,483
• ai1,955
• linux4,184
• open-source5,666
• macos3,544
• self-hosted1,929
• windows3,403
• python2,383
• ollama236
• docker1,660
• cli2,227
• mit3,312
• cross-platform2,623
• gpu789
• server1,459
• ai-agents1,289
• privacy886
• inference327
• openai576
• openai-compatible211
• web2,447
• apache-2.01,434
• mcp913
• developer-tools2,924