[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1768":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":9,"totalLinesOfCode":9,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":9,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":9,"rankLanguage":9,"license":9,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":21,"topics":23,"createdAt":9,"pushedAt":9,"updatedAt":38,"readmeContent":39,"aiSummary":40,"trendingCount":16,"starSnapshotCount":16,"syncStatus":41,"lastSyncTime":42,"discoverSource":43},1768,"nexa-sdk","qualcomm\u002Fnexa-sdk","qualcomm","Run frontier LLMs and VLMs with day-0 model support across GPU, NPU, and CPU, with comprehensive runtime coverage for PC (Python\u002FC++), mobile (Android & iOS), and Linux\u002FIoT (Arm64 & x86 Docker). Supporting OpenAI GPT-OSS, IBM Granite-4, Qwen-3-VL, Gemma-3n, Ministral-3, and more.",null,"https:\u002F\u002Fgithub.com\u002Fqualcomm\u002Fnexa-sdk","Rust",8094,1002,527,48,0,13,49,5,73.4,false,"main",[24,25,26,27,28,29,30,31,32,33,34,35,36,37],"llm","on-device-ai","sdk","stable-diffusion","vlm","go","gemma3","gpt-oss","llama","llama3","phi3","qwen3","granite4","qwen3vl","2026-06-12 04:00:11","\u003Cdiv align=\"center\" style=\"text-decoration: none;\">\n  \u003Cimg width=\"100%\" src=\"assets\u002Fbanner1.png\" alt=\"Nexa AI Banner\">\n  \u003Cp>\n    \u003Ca href=\"https:\u002F\u002Fwww.producthunt.com\u002Fproducts\u002Fnexasdk-for-mobile?embed=true&utm_source=badge-top-post-badge&utm_medium=badge&utm_campaign=badge-nexasdk-for-mobile\" target=\"_blank\" rel=\"noopener noreferrer\">\n        \u003Cimg alt=\"NexaSDK for Mobile - #1 Product of the Day\" width=\"180\" height=\"39\" src=\"https:\u002F\u002Fapi.producthunt.com\u002Fwidgets\u002Fembed-image\u002Fv1\u002Ftop-post-badge.svg?post_id=1049998&theme=dark&period=daily&t=1765991451976\">\n    \u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Ftrendshift.io\u002Frepositories\u002F12239\" target=\"_blank\" rel=\"noopener noreferrer\">\n        \u003Cimg alt=\"NexaAI\u002Fnexa-sdk - #1 Repository of the Day\" height=\"39\" src=\"https:\u002F\u002Ftrendshift.io\u002Fapi\u002Fbadge\u002Frepositories\u002F12239\">\n    \u003C\u002Fa>\n  \u003C\u002Fp>\n  \u003Cp>\n    \u003Ca href=\"https:\u002F\u002Fdocs.nexa.ai\">\n        \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdocs-website-brightgreen?logo=readthedocs\" alt=\"Documentation\">\n    \u003C\u002Fa>\n  \u003C\u002Fp>\n\u003C\u002Fdiv>\n\n# NexaSDK\n\n**NexaSDK lets you build the smartest and fastest on-device AI with minimum energy.** It is a highly performant local inference framework that runs the latest multimodal AI models locally on NPU, GPU, and CPU - across Android, Windows, and Linux devices with a few lines of code.\n\nNexaSDK supported latest models **weeks or months before anyone else** — Qwen3-VL, DeepSeek-OCR, Gemma3n (Vision), and more.\n\n> ⭐ **Star this repo** to keep up with exciting updates and new releases about latest on-device AI capabilities.\n\n## 🏆 Recognized Milestones\n\n- **Qualcomm** featured us **3 times** in official blogs.\n  - [Innovating Multimodal AI on Qualcomm Hexagon NPU](https:\u002F\u002Fwww.qualcomm.com\u002Fdeveloper\u002Fblog\u002F2025\u002F09\u002Fomnineural-4b-nexaml-qualcomm-hexagon-npu).\n  - [First-ever Day-0 model support on Qualcomm Hexagon NPU for compute and mobile platforms, Auto and IoT](https:\u002F\u002Fwww.qualcomm.com\u002Fdeveloper\u002Fblog\u002F2025\u002F10\u002Fgranite-4-0-to-the-edge-on-device-ai-for-real-world-performance).\n  - [A simple way to bring on-device AI to smartphones with Snapdragon](https:\u002F\u002Fwww.qualcomm.com\u002Fdeveloper\u002Fblog\u002F2025\u002F11\u002Fnexa-ai-for-android-simple-way-to-bring-on-device-ai-to-smartphones-with-snapdragon)\n\n## 🚀 Quick Start\n\n| Platform        | Links                                                                                     |\n| --------------- | ----------------------------------------------------------------------------------------- |\n| 🖥️ CLI          | [Quick Start](#-cli) ｜ [Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-go\u002FNexaCLI)               |\n| 🐍 Python       | [Quick Start](#-python-sdk) ｜ [Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-python\u002Foverview)   |\n| 🤖 Android      | [Quick Start](#-android-sdk) ｜ [Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-android\u002Foverview) |\n| 🐳 Linux Docker | [Quick Start](#-linux-docker) ｜ [Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-docker\u002Foverview) |\n\n---\n\n### 🖥️ CLI\n\n**Download:**\n\n| Windows                                                                                                  | Linux                                                                                        |\n| -------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- |\n| [arm64 (Qualcomm NPU)](https:\u002F\u002Fpublic-storage.nexa4ai.com\u002Fnexa_sdk\u002Fdownloads\u002Fnexa-cli_windows_arm64.exe) | [arm64](https:\u002F\u002Fgithub.com\u002FNexaAI\u002Fnexa-sdk\u002Freleases\u002Flatest\u002Fdownload\u002Fnexa-cli_linux_arm64.sh) |\n| [x64](https:\u002F\u002Fpublic-storage.nexa4ai.com\u002Fnexa_sdk\u002Fdownloads\u002Fnexa-cli_windows_x86_64.exe) | [x64](https:\u002F\u002Fgithub.com\u002FNexaAI\u002Fnexa-sdk\u002Freleases\u002Flatest\u002Fdownload\u002Fnexa-cli_linux_x86_64.sh)  |\n\n\n\n**NPU Access Token (required for NPU models):**\n\n> **Note:** Our previous token validation service has been deprecated. For any NPU usage, simply set the access token below — no additional registration or validation is needed.\n\nFor Windows:\n```shell\n$env:NEXA_TOKEN=\"key\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.BXoUHIEzFMuuZbBT7RvsKO9nTi5950C6kHO64blF7XBnfKvZ6ClA8a55tmszI1ZWdngzpNFTzMM5PV5euuzMCA==\"\n```\n\nFor Linux \u002F Android adb shell:\n```shell\nexport NEXA_TOKEN=\"key\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.BXoUHIEzFMuuZbBT7RvsKO9nTi5950C6kHO64blF7XBnfKvZ6ClA8a55tmszI1ZWdngzpNFTzMM5PV5euuzMCA==\"\n```\n\n**Run your first model:**\n\n```bash\n# Chat with Qwen3\nnexa infer ggml-org\u002FQwen3-1.7B-GGUF\n\n# Multimodal: drag images into the CLI\nnexa infer NexaAI\u002FQwen3-VL-4B-Instruct-GGUF\n\n# NPU (Windows arm64 with Snapdragon X Elite)\nnexa infer NexaAI\u002FOmniNeural-4B\n```\n\n- **Models:** LLM, Multimodal, ASR, OCR, Rerank, Object Detection, Image Generation, Embedding\n- **Formats:** GGUF, NEXA\n- 📖 [CLI Reference Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-go\u002FNexaCLI)\n\n---\n\n### 🐍 Python SDK\n\n```bash\npip install nexaai\n```\n\n```python\nfrom nexaai import LLM, GenerationConfig, ModelConfig, LlmChatMessage\n\nllm = LLM.from_(model=\"NexaAI\u002FQwen3-0.6B-GGUF\", config=ModelConfig())\n\nconversation = [\n    LlmChatMessage(role=\"user\", content=\"Hello, tell me a joke\")\n]\nprompt = llm.apply_chat_template(conversation)\nfor token in llm.generate_stream(prompt, GenerationConfig(max_tokens=100)):\n    print(token, end=\"\", flush=True)\n```\n\n- **Models:** LLM, Multimodal, ASR, OCR, Rerank, Object Detection, Image Generation, Embedding\n- **Formats:** GGUF, NEXA\n- 📖 [Python SDK Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-python\u002Fquickstart)\n\n---\n\n### 🤖 Android SDK\n\nAdd to your `app\u002FAndroidManifest.xml`\n\n```xml\n\u003Capplication android:extractNativeLibs=\"true\">\n```\n\nAdd to your `build.gradle.kts`:\n\n```kotlin\ndependencies {\n    implementation(\"ai.nexa:core:0.0.19\")\n}\n```\n\n```kotlin\n\u002F\u002F Initialize SDK\nNexaSdk.getInstance().init(this)\n\n\u002F\u002F Load and run model\nVlmWrapper.builder()\n    .vlmCreateInput(VlmCreateInput(\n        model_name = \"omni-neural\",\n        model_path = \"\u002Fdata\u002Fdata\u002Fyour.app\u002Ffiles\u002Fmodels\u002FOmniNeural-4B\u002Ffiles-1-1.nexa\",\n        plugin_id = \"npu\",\n        config = ModelConfig()\n    ))\n    .build()\n    .onSuccess { vlm ->\n        vlm.generateStreamFlow(\"Hello!\", GenerationConfig()).collect { print(it) }\n    }\n```\n\n- **Requirements:** Android minSdk 27, Qualcomm Snapdragon 8 Gen 4 Chip\n- **Models:** LLM, Multimodal, ASR, OCR, Rerank, Embedding\n- **NPU Models:** [Supported Models](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-android\u002Foverview#supported-models)\n- 📖 [Android SDK Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-android\u002Fquickstart)\n\n---\n\n### 🐳 Linux Docker\n\n```bash\ndocker pull nexa4ai\u002Fnexasdk:latest\n\nexport NEXA_TOKEN=\"your_token_here\"\ndocker run --rm -it --privileged \\\n  -e NEXA_TOKEN \\\n  nexa4ai\u002Fnexasdk:latest infer NexaAI\u002FGranite-4.0-h-350M-NPU\n```\n\n- **Requirements:** Qualcomm Dragonwing IQ9, ARM64 systems\n- **Models:** LLM, VLM, ASR, CV, Rerank, Embedding\n- **NPU Models:** [Supported Models](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-docker\u002Foverview#supported-models)\n- 📖 [Linux Docker Docs](https:\u002F\u002Fdocs.nexa.ai\u002Fen\u002Fnexa-sdk-docker\u002Fquickstart)\n\n---\n\n## ⚙️ Features & Comparisons\n\n\u003Cdiv align=\"center\">\n\n| Features                                 | **NexaSDK**                                                | **Ollama** | **llama.cpp** | **LM Studio** |\n| ---------------------------------------- | ---------------------------------------------------------- | ---------- | ------------- | ------------- |\n| NPU support                              | ✅ NPU-first                                               | ❌         | ❌            | ❌            |\n| Android SDK support                  | ✅ NPU\u002FGPU\u002FCPU support                                     | ⚠️         | ⚠️            | ❌            |\n| Linux support (Docker image)             | ✅                                                         | ✅         | ✅            | ❌            |\n| Day-0 model support  | ✅                                                         | ❌         | ⚠️            | ❌            |\n| Full multimodality support               | ✅ Image, Audio, Text, Embedding, Rerank, ASR, TTS         | ⚠️         | ⚠️            | ⚠️            |\n| Cross-platform support                   | ✅ Desktop, Mobile (Android), Automotive, IoT (Linux) | ⚠️         | ⚠️            | ⚠️            |\n| One line of code to run                  | ✅                                                         | ✅         | ⚠️            | ✅            |\n| OpenAI-compatible API + Function calling | ✅                                                         | ✅         | ✅            | ✅            |\n\n\u003Cp align=\"center\" style=\"margin-top:14px\">\n  \u003Ci>\n      \u003Cb>Legend:\u003C\u002Fb>\n      \u003Cspan title=\"Full support\">✅ Supported\u003C\u002Fspan>   |  \n      \u003Cspan title=\"Partial or limited support\">⚠️ Partial or limited support \u003C\u002Fspan>   |  \n      \u003Cspan title=\"Not Supported\">❌ No\u003C\u002Fspan>\n  \u003C\u002Fi>\n\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n## 🙏 Acknowledgements\n\nWe would like to thank the following projects:\n\n- [ggml](https:\u002F\u002Fgithub.com\u002Fggml-org\u002Fggml)\n- [mlx-lm](https:\u002F\u002Fgithub.com\u002Fml-explore\u002Fmlx-lm)\n- [mlx-vlm](https:\u002F\u002Fgithub.com\u002FBlaizzy\u002Fmlx-vlm)\n- [mlx-audio](https:\u002F\u002Fgithub.com\u002FBlaizzy\u002Fmlx-audio)\n\n## 📄 License\n\nNexaSDK uses a dual licensing model:\n\n### CPU\u002FGPU Components\n\nLicensed under [Apache License 2.0](LICENSE).\n\n### NPU Components\n\n- **Personal Use**: Free license key available from [Nexa AI Model Hub](https:\u002F\u002Fsdk.nexa.ai\u002Fmodel). Each key activates 1 device for NPU usage.\n- **Commercial Use**: Contact [hello@nexa.ai](mailto:hello@nexa.ai) for licensing.\n\n## 🤝 Contact & Community Support\n\nWant more model support, backend support, device support or other features? We'd love to hear from you!\n\nFeel free to [submit an issue](https:\u002F\u002Fgithub.com\u002FNexaAI\u002Fnexa-sdk\u002Fissues) on our GitHub repository with your requests, suggestions, or feedback. Your input helps us prioritize what to build next.\n","NexaSDK 是一个高性能的本地推理框架，支持在GPU、NPU和CPU上运行最新的多模态AI模型，适用于PC（Python\u002FC++）、移动设备（Android & iOS）以及Linux\u002FIoT（Arm64 & x86 Docker）。其核心功能包括对前沿的大规模语言模型（LLM）和视觉语言模型（VLM）的支持，如Qwen-3-VL、Gemma-3n等，并且能在这些模型公开后的几周或几个月内提供支持。此外，NexaSDK通过优化能源消耗实现了高效的本地AI处理。此项目特别适合需要在多种硬件平台上快速部署最新AI模型的应用场景，例如智能手机应用开发、边缘计算解决方案和个人电脑上的AI服务。",2,"2026-06-11 02:45:54","trending"]