{"data":{"slug":"alibaba-mnn","name":"MNN","tagline":"Blazing-fast, lightweight inference engine for high-performance on-device LLMs and Edge AI","github_url":"https://github.com/alibaba/MNN","owner":"alibaba","repo":"MNN","owner_avatar_url":"https://avatars.githubusercontent.com/u/1961952?v=4","primary_language":"C++","stars":15632,"forks":2383,"topics":["arm","convolution","deep-learning","embedded-devices","llm","machine-learning","ml","mnn","transformer","vulkan","winograd-algorithm"],"archived":false,"github_pushed_at":"2026-07-09T09:50:18+00:00","maintenance_label":"Very active","url":"https://www.graphcanon.com/tools/alibaba-mnn","markdown_url":"https://www.graphcanon.com/tools/alibaba-mnn.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/alibaba-mnn","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=alibaba-mnn","description":"MNN: A blazing-fast, lightweight inference engine battle-tested by Alibaba, powering high-performance on-device LLMs and Edge AI.","homepage_url":null,"license":"Apache-2.0","open_issues":49,"watchers":237,"ai_summary":"MNN is a powerful and efficient inference engine designed for deploying deep learning models at the edge, facilitating real-time performance on various devices.","readme_excerpt":"---\n\n\n\n\n\n\n\n\n\n\n\n\n## News 🔥\n- [2026/03/05] Support Qwen3.5 Series.\n
\n \n \n \n
\n\n- [2026/02/13] MNN-Sana-Edit-V2 is now available at [apps](./apps/sana/README.md), offering cartoon-style photo editing based on Sana.\n
\n \n
\n\n\n History News \n\n- [2025/10/16] Support Qwen3-VL Series.\n- [2025/06/11] New App MNN TaoAvatar released, you can talk with 3DAvatar offline with LLM, ASR, TTS, A2BS and NNR models all run local on your device!! [MNN TaoAvatar](./apps/Android/Mnn3dAvatar/README.md)\n
\n \n
\n\n- [2025/05/12] android app support qwen2.5 omni 3b and 7b [MNN Chat App](./apps/Android/MnnLlmChat/README.md#releases).\n
\n \n \n \n
\n\n- [2025/04/30] android app support qwen3 and dark mode [MNN Chat App](./apps/Android/MnnLlmChat/README.md#releases).\n
\n \n
\n\n- [2025/02/18] iOS multimodal LLM App is released [MNN LLM iOS](./apps/iOS/MNNLLMChat/README.md).\n
\n \n
\n\n- [2025/02/11] android app support for [deepseek r1 1.5b](./project/android/apps/MnnLlmApp/README.md#version-021).\n
\n \n
\n\n- [2025/01/23] We released our full multimodal LLM Android App:[MNN-LLM-Android](./apps/Android/MnnLlmChat/README.md). including text-to-text, image-to-text, audio-to-text, and text-to-image generation.\n
\n \n \n \n \n
\n\n\n## Intro\nMNN is a highly efficient and lightweight deep learning framework. It supports inference and training of deep learning models and has industry-leading performance for inference and training on-device. At present, MNN has been integrated into more than 30 apps of Alibaba Inc, such as Taobao, Tmall, Youku, DingTalk, Xianyu, etc., covering more than 70 usage scenarios such as live broadcast, short video capture, search recommendation, product searching by image, interactive marketing, equity distribution, security risk control. In addition, MNN is also used on embedded devices, such as IoT.\n\n[MNN-LLM](./transformers/README.md) is a large language model runtime solution developed based on the MNN engine. The mission of this project is to deploy LLM models locally on everyone's platforms(Mobile Phone/PC/IOT). It supports popular large language models such as Qianwen, Baichuan, Zhipu, LLAMA, and others. [MNN-LLM","github_created_at":"2019-04-15T07:40:18+00:00","created_at":"2026-07-11T10:40:10.53019+00:00","updated_at":"2026-07-11T14:53:13.444924+00:00","categories":[{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"}],"tags":[{"slug":"ml","name":"ml"},{"slug":"convolution","name":"convolution"},{"slug":"deep-learning","name":"deep-learning"},{"slug":"arm","name":"arm"},{"slug":"llm","name":"llm"},{"slug":"embedded-devices","name":"embedded-devices"},{"slug":"machine-learning","name":"machine-learning"},{"slug":"transformer","name":"transformer"}],"trust":{"provenance":{"is_fork":false,"github_id":181436799,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-11T10:40:11.220Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":1,"days_since_push":2,"last_release_at":"2026-06-16T07:58:06Z"},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T10:40:11.915Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-07-11T14:52:31.485Z"},"languages":{"value":["c++"],"source":"github.language","observed_at":"2026-07-11T14:52:31.485Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-07-11T14:52:31.485Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":{"min_ram_gb":2,"requires_docker":false},"constraints":{"min_ram_gb":2,"requires_docker":false},"when_to_use":["- When you need lightning-fast and low-memory usage performance on mobile devices or edge computing environments.","- If your application needs to run locally without cloud dependency, maximizing privacy and reducing latency.","- You are working with the Qwen series of models (e.g., Qwen3.5) as MNN is battle-tested for high-performance on-device inference.","- When deploying applications that require real-time performance in scenarios like live broadcasts or short video capture."],"when_not_to_use":["- If your primary requirement is training deep learning models, since MNN mainly focuses on fast and lightweight inference rather than heavy-duty training tasks.","- For applications requiring significant external data access or continuous cloud updates, as MNN emphasizes local processing.","- When you are developing for platforms that require non-native support; MNN is optimized for native integration with Alibaba's ecosystem but might not offer the same level of support for other third-"],"source":"enrich:decision_facts","observed_at":"2026-07-11T14:53:13.106Z"},"constraint_facets":{"min_ram_gb":2,"requires_docker":false},"decision_summary":[{"label":"Requirements","value":"Min 2 GB RAM"},{"label":"Adopt for","value":"MNN is a highly efficient and lightweight deep learning framework designed for high-performance inference on-device. Developed by Alibaba, it supports various applications across multiple Alibaba platforms."},{"label":"License detail","value":"MNN is licensed under Apache-2.0, allowing free use and modification in both community projects and commercial applications."}]}}
![\"Icon\"](\"https://meta.alicdn.com/data/mnn/assets/qwen35_1.jpg\")
\n ![\"Icon\"](\"https://meta.alicdn.com/data/mnn/assets/qwen35_2.jpg\")
\n ![\"Icon\"](\"https://meta.alicdn.com/data/mnn/assets/qwen35_3.jpg\")
\n![\"Icon\"](\"https://meta.alicdn.com/data/mnn/assets/sana_show_case.jpg\")
\n![\"Icon\"](\"https://meta.alicdn.com/data/mnn/avatar/avatar_demo.gif\"){kind=avatar}
\n![\"Icon\"](\"./apps/Android/MnnLlmChat/assets/image_home_new.jpg\")
\n ![\"Icon\"](\"./apps/Android/MnnLlmChat/assets/image_sound_new.jpg\")
\n ![\"Icon\"](\"./apps/Android/MnnLlmChat/assets/image_image_new.jpg\")
\n![\"Icon\"](\"https://meta.alicdn.com/data/mnn/qwen_3.gif\")
\n![\"Icon\"](\"./apps/iOS/MNNLLMChat/assets/introduction.gif\")
\n![\"Icon\"](\"./apps/Android/MnnLlmChat/assets/deepseek_support.gif\")
\n![\"Icon\"](\"./apps/Android/MnnLlmChat/assets/image_diffusion_new.jpg\")
\n