Local & Open Models
Running, serving and fine-tuning open-weight models. As of 2026-11-21, 388 indexed products match this subcategory (230 with full profiles); 115 were published in the last 90 days, against 85 in the 90 days before, slower than the catalogue as a whole. By GitHub stars, cherry-studio leads, holding 10% of the stars of the 331 products here with their own repository. The most common form is MCP servers (64% of indexed products).
SubcategorySegment label: Lagging the catalogue DER
Segment statistics
Whole catalogue (all 388 indexed products, not only profiled ones). Calculated 2026-11-21.
Lagging the catalogue. Mid-sized for its level, with launches growing more slowly than the catalogue’s. Size is ranked against the other subcategory nodes; momentum compares this node’s launches in the last 90 days with the 90 before, relative to the whole catalogue’s ratio (score 62 or more = faster, 38 or less = slower; below 5 dated launches it is not measured and the label uses size alone).
Concentration. The Herfindahl–Hirschman index (HHI) adds up each project’s squared share of stars. At 0.042, attention is spread as if across about 24 equally popular projects. Using the common competition-policy thresholds (0.15 and 0.25), that is unconcentrated. Stars measure attention, not usage or quality.
Which product forms serve it
The same need is met by MCP servers, agent skills, plugins, extensions and frameworks. This counts each form across the whole catalogue.
| Product type | Indexed | Share | With a profile |
|---|---|---|---|
| MCP servers | 249 | 64.2% | 96 |
| Agent skills | 43 | 11.1% | 42 |
| Infrastructure | 35 | 9.0% | 32 |
| IDE extensions | 17 | 4.4% | 18 |
| Frameworks | 14 | 3.6% | 13 |
| Agent projects | 13 | 3.4% | 13 |
| Plugins | 10 | 2.6% | 9 |
| Libraries | 3 | 0.77% | 3 |
| AI APIs | 2 | 0.52% | 2 |
| Rules packs | 1 | 0.26% | 1 |
| Developer tools | 1 | 0.26% | 1 |
3 capabilities
Indexed = whole catalogue; profiled = products with their own page. A product can sit in several, so the rows do not add up to the total.
| Capability | Indexed | Profiled | New, 90 days | Label |
|---|---|---|---|---|
| Local InferenceRunning models on your own machine. | 256 | 150 | 89 | Crowded, slowing |
| Model Fine-tuningTraining and adapting models on custom data. | 64 | 41 | 8 | Lagging the catalogue |
| Model ServingInference servers for production model hosting. | 54 | 31 | 17 | Lagging the catalogue |
Top 40 products by GitHub stars
Search the directoryProfiled products only (230 of 388), ranked by the stars of their own repository. Products that live in a subfolder of a larger repository (common for skills) show that whole repository's stars and are ranked after them; products without a repository come last. Stars are popularity, not quality or installs.
| # | Product | Type | Publisher | Capabilities | GitHub stars | Last push | Licence | Save and compare |
|---|---|---|---|---|---|---|---|---|
| 1 | cherry-studioYour AI workspace for models, agents, and productivity. | Agent skill | CherryHQ | 54k | today | AGPL-3.0 | ||
| 2 | nanobotYour lightweight open-source personal agent. | MCP server | HKUDS | 50k | today | MIT | ||
| 3 | CodewhaleAn open-source coding agent for your terminal, written in Rust. It works with any model, can split one task across several agents, and every line of it is yours to read. | Agent project | codewhale-hq | 41k | today | MIT | ||
| 4 | distillyDistilly — Distill how they think into reusable Skills for any Agent or Bot. Formerly Colleague Skill(原同事 Skill). | Agent skill | titanwings | 27k | 1mo ago | MIT | ||
| 5 | MaxKB🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。 | MCP server | 1Panel-dev | 23k | today | GPL-3.0 | ||
| 6 | awesome-ai-engineering🚀 A curated collection of AI engineering resources, roadmaps, tools, and projects covering LLMs, RAG, AI Agents, MCP, fine-tuning, evaluations, and production AI systems. | Agent project | dipakkr | 23k | 1mo ago | MIT | ||
| 7 | docker-androidAndroid in docker solution with noVNC supported, video recording, mcp server and AI-agent | MCP server | budtmo | 16k | 2d ago | — | ||
| 8 | AI-Research-SKILLsComprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepower. Maintained by Orchestra Research. | Agent skill | Orchestra-Research | 15k | 1mo ago | MIT | ||
| 9 | BrowserOS🌐 The open-source Agentic browser; alternative to ChatGPT Atlas, Perplexity Comet, Dia. | Agent project | browseros-ai | 14k | today | AGPL-3.0 | ||
| 10 | LEANN[MLsys2026 Best Paper]: https://arxiv.org/abs/2506.08276. RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device. | Infrastructure | StarTrail-org | 14k | 1d ago | MIT | ||
| 11 | langchain4jLangChain4j is an idiomatic, open-source Java library for building LLM-powered applications on the JVM. It offers a unified API over popular LLM providers and vector stores, and makes implementing tool calling (including MCP support), agents and RAG easy. It integrates seamlessly with enterprise Java frameworks like Quarkus and Spring Boot. | Infrastructure | langchain4j | 13k | today | Apache-2.0 | ||
| 12 | SoupFine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU. | Infrastructure | MakazhanAlpamys | 11k | today | Apache-2.0 | ||
| 13 | aichatAll-in-one LLM CLI tool featuring Shell Assistant, Chat-REPL, RAG, AI Tools & Agents, with access to OpenAI, Claude, Gemini, Ollama, Groq, and more. | Agent project | sigoden | 11k | 9mo ago | Apache-2.0 | ||
| 14 | oumiEasily fine-tune (SFT/RL), evaluate, and deploy Qwen, Gemma, or any open agentic LLM/VLM! | Infrastructure | oumi-ai | 9.4k | today | Apache-2.0 | ||
| 15 | BentoMLThe easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more! | Infrastructure | bentoml | 8.9k | 4d ago | Apache-2.0 | ||
| 16 | osaurusOwn your AI. The open-source agent harness for macOS. Run local or cloud models with persistent memory, skills, tools, and autonomous execution. Built in Swift. | MCP server | osaurus-ai | 8.2k | today | MIT | ||
| 17 | big-AGIThe AI workspace for experts. Switch across 800+ models mid-chat, tune 40+ parameters, Beam models side by side. BYO keys, self-hostable. Built on our native-protocol engine: reliable, hyper-configurable, and fast. | Agent project | enricoros | 7.2k | 1d ago | MIT | ||
| 18 | open-multi-agentSelf-hosted TypeScript agent runtime with durable approvals and verifiable run records. Own it, approve it, audit it. | Framework | open-multi-agent | 7.1k | 4d ago | MIT | ||
| 19 | bla🛡️ BLA (bla-agents / Blind LLM Agents) — Open-source Blue Defense MCP server & CLI protecting JS/TS, Electron ASAR, WASM, Python, Native Binaries, SQLite, GGUF/Safetensors & Android DEX against AI reverse-engineering agents (REA). | MCP server | qybaihe | 6.4k | today | MIT | ||
| 20 | AgentOpsSDK and dashboard for monitoring agent sessions, tool calls, costs and failures, with integrations for CrewAI, OpenAI Agents SDK, AG2 and others. | Developer tool | AgentOps | 6.0k | 4mo ago | MIT | ||
| 21 | OlaresOpen-Source Personal Cloud OS for Always-On Agents | Agent project | beclab | 5.4k | today | AGPL-3.0 | ||
| 22 | cube-studiocube studio开源云原生一站式机器学习/深度学习/大模型AI平台,mlops算法链路全流程,算力租赁平台,notebook在线开发,拖拉拽任务流pipeline编排,多机多卡分布式训练,超参搜索,推理服务VGPU虚拟化,边缘计算,标注平台自动化标注,deepseek等大模型sft微调/奖励模型/强化学习训练,vllm/ollama/mindie大模型多机推理,私有知识库,AI模型市场,支持国产cpu/gpu/npu 昇腾生态,支持RDMA,支持pytorch/tf/mxnet/deepspeed/paddle/colossalai/horovod/ray/volcano等分布式 | Infrastructure | tencentmusic | 5.1k | 1mo ago | — | ||
| 23 | deep-researchUse any LLMs (Large Language Models) for Deep Research. Support SSE API and MCP server. | MCP server | u14app | 4.7k | 5mo ago | MIT | ||
| 24 | koogKoog is a JVM (Java and Kotlin) framework for building predictable, fault-tolerant and enterprise-ready AI agents across all platforms – from backend services to Android and iOS, JVM, and even in-browser environments. Koog is based on our AI products expertise and provides proven solutions for complex LLM and AI problems | Framework | JetBrains | 4.7k | 11d ago | Apache-2.0 | ||
| 25 | LongMemoryLocal persistent memory store for LLM applications including claude desktop, github copilot, codex, antigravity, etc. | Agent project | CaviraOSS | 4.6k | 11d ago | Apache-2.0 | ||
| 26 | locally-uncensoredThe all-in-one local AI studio for your desktop: chat, image and video generation and a coding agent in one free, open source app. Windows and Linux. No Docker, no terminal, no cloud required. | Agent project | PurpleDoubleD | 4.5k | 2d ago | AGPL-3.0 | ||
| 27 | claude-octopusRun multiple AI models against the same research, design, or coding task. Surface disagreements before you ship. | Plugin | nyldn | 4.5k | 2d ago | MIT | ||
| 28 | octov11.26.0 - Custom provider presets in the tool helper, native skills and debate advisor checks. 31 personas, 56 commands, 64 skills. Run /octo:setup. | Plugin | nyldn | 4.5k | 2d ago | MIT | ||
| 29 | agentic-rag-for-dummiesA modular Agentic RAG built with LangGraph — learn Retrieval-Augmented Generation Agents in minutes. | Agent project | GiovanniPasq | 4.4k | 11d ago | MIT | ||
| 30 | langroidHarness LLMs with Multi-Agent Programming | Framework | langroid | 4.2k | today | MIT | ||
| 31 | pipeshub-aiThe open-source context layer for AI agents and enterprise AI search. Permission-aware search and cited answers across 30+ sources, including Slack, Drive, Jira, GitHub and Microsoft 365, for people and agents through MCP, SDKs and built-in agents. Self-hosted or Cloud. | MCP server | pipeshub-ai | 4.1k | today | Apache-2.0 | ||
| 32 | twinny - AI Code Completion and ChatAI code completion, chat, inline edit and code review on models you run. Private, open source, one gateway for the whole team. | IDE extension | twinnydotdev | 3.7k | 3d ago | MIT | ||
| 33 | chat-ollamaChatOllama is an open source agentic app for running AI agents across local and hosted models. | MCP server | sugarforever | 3.5k | 3d ago | — | ||
| 34 | second-brain-ai-assistant-courseLearn to build your Second Brain AI assistant with LLMs, agents, RAG, fine-tuning, LLMOps and AI systems techniques. | Infrastructure | itsdecodingai | 3.1k | 7mo ago | MIT | ||
| 35 | vllm-ascendCommunity maintained hardware plugin for vLLM on Huawei Ascend | Infrastructure | vllm-project | 3.1k | today | Apache-2.0 | ||
| 36 | pinvou-agentOpen-source desktop AI agent for tools, files, knowledge, workflows, and real deliverables. | Agent skill | Pinvou | 3.0k | today | MIT | ||
| 37 | little-coderA harness optimized to smaller LLMs | Agent project | itayinbarr | 2.7k | 2mo ago | Apache-2.0 | ||
| 38 | cube-studiocubestudio开源云原生一站式机器学习/深度学习/大模型AI平台/MaaS/mlops/人工智能平台/训推平台,算法全链路流程,多租户,算力租赁平台,token中转,拖拉拽任务流pipeline编排,多机多卡分布式训练,超参搜索,推理服务,VGPU虚拟化,云边端协同,边缘计算,自动化标注平台,deepseek等大模型sft微调/奖励模型/强化学习训练,vllm/ollama/mindie大模型多机推理,私有知识库llmops智能体,AI模型市场,支持国产异构算力调度,昇腾/寒武纪/海光/摩尔/沐曦等,支持ib/roce/RDMA,信创支持 | Infrastructure | data-infra | 2.6k | 1mo ago | — | ||
| 39 | agency-orchestrator🚀 One sentence → your one-person company of AI experts → complete deliverable in minutes. 276 CN + 184 EN + 5 more languages (ko/ru/pt-BR/id/ar) · zero-code YAML · auto-verified acceptance · Web Studio / Desktop / Docker · 15 LLM providers (11 key-free). 一句话组建你的「一人公司」AI 专家团队,几分钟交付完整方案;验收自动核验,网页 / 桌面 / Docker 全渠道。 | MCP server | jnMetaCode | 2.5k | 10d ago | Apache-2.0 | ||
| 40 | envd🏕️ Reproducible development environment for humans and agents | Infrastructure | tensorchord | 2.2k | 2d ago | Apache-2.0 |
Most common integrations
Third-party services named by products in this node, whole catalogue.
| Integration | Products | Share |
|---|---|---|
| GitHub | 5 | 1.3% |
| Android | 4 | 1.0% |
| Blender | 4 | 1.0% |
| Obsidian | 4 | 1.0% |
| Azure | 2 | 0.5% |
| Gmail | 2 | 0.5% |
| 2 | 0.5% | |
| PostgreSQL | 2 | 0.5% |
| Confluence | 1 | 0.3% |
| Databricks | 1 | 0.3% |
| Google Drive | 1 | 0.3% |
| Grafana | 1 | 0.3% |
| Jira | 1 | 0.3% |
| Microsoft Excel | 1 | 0.3% |
| Neo4j | 1 | 0.3% |
Launches per month
New Local & Open Models products per monthSRC
Whole catalogue, by publication month, 2024-11 to 2026-11. The last month is partial; undated products are excluded.
Show the data as a table
| Products published (products) | |
|---|---|
| 2024-11 | 2 |
| 2024-12 | 2 |
| 2025-01 | 4 |
| 2025-02 | 5 |
| 2025-03 | 1 |
| 2025-04 | 2 |
| 2025-05 | 5 |
| 2025-06 | 1 |
| 2025-07 | 1 |
| 2025-08 | 2 |
| 2025-10 | 3 |
| 2025-11 | 3 |
| 2025-12 | 3 |
| 2026-01 | 6 |
| 2026-02 | 15 |
| 2026-03 | 20 |
| 2026-04 | 21 |
| 2026-05 | 13 |
| 2026-06 | 20 |
| 2026-07 | 39 |
| 2026-08 | 27 |
| 2026-09 | 28 |
| 2026-10 | 54 |
| 2026-11 | 29 |
Related opportunities
Opportunity radarScored gaps for this node or anything beneath it. Scores rank hypotheses; they do not predict demand.
| Opportunity | Competition | Confidence | Score |
|---|---|---|---|
| Coverage gap: Model ServingModel Serving | High | Low | 52.1 |
Limits of these numbers
- Products are placed here by whole-word keyword rules on their name, description and repository topics, and inherit every ancestor of a matched node. Products that do this without naming it are missed; ambiguous words cause false matches. Methodology
- About half of the catalogue matches no node at all, so absolute counts understate every category.
- Launch counts use the publication date each source reports; for auto-discovered repositories that is the repository’s creation date.
- Stars and concentration cover only products with their own repository (not a subfolder of a larger one). They measure attention, not adoption.
Page data last updated .