{"id":29182342,"url":"https://github.com/openvinotoolkit/openvino_testdrive","last_synced_at":"2025-07-01T20:05:26.993Z","repository":{"id":261420187,"uuid":"863449409","full_name":"openvinotoolkit/openvino_testdrive","owner":"openvinotoolkit","description":"With OpenVINO Test Drive, users can run large language models (LLMs) and models trained by Intel Geti on their devices, including AI PCs and Edge devices.","archived":false,"fork":false,"pushed_at":"2025-05-14T14:55:54.000Z","size":25348,"stargazers_count":25,"open_issues_count":5,"forks_count":2,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-05-14T15:51:08.755Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Dart","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/openvinotoolkit.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":".github/CODEOWNERS","security":"security.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-09-26T10:08:30.000Z","updated_at":"2025-05-14T08:58:27.000Z","dependencies_parsed_at":"2024-11-06T13:48:13.950Z","dependency_job_id":"c4c0f3e5-c358-4333-a9ac-1ac88890be66","html_url":"https://github.com/openvinotoolkit/openvino_testdrive","commit_stats":null,"previous_names":["openvinotoolkit/openvino_testdrive"],"tags_count":3,"template":false,"template_full_name":null,"purl":"pkg:github/openvinotoolkit/openvino_testdrive","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openvinotoolkit%2Fopenvino_testdrive","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openvinotoolkit%2Fopenvino_testdrive/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openvinotoolkit%2Fopenvino_testdrive/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openvinotoolkit%2Fopenvino_testdrive/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/openvinotoolkit","download_url":"https://codeload.github.com/openvinotoolkit/openvino_testdrive/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openvinotoolkit%2Fopenvino_testdrive/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":263029207,"owners_count":23402354,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-07-01T20:03:45.114Z","updated_at":"2025-07-01T20:05:26.946Z","avatar_url":"https://github.com/openvinotoolkit.png","language":"Dart","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n# OpenVINO™ Test Drive\n\n[![openvino](https://img.shields.io/badge/openvino-2025.0-blue)]()\n\n\u003c/div\u003e\n\nGet started with OpenVINO™ Test Drive, an application that allows you to run generative AI and vision models trained by [Intel® Geti™](https://docs.geti.intel.com/) directly on your computer or edge device using [OpenVINO™ Runtime](https://github.com/openvinotoolkit/openvino).\n\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./docs/llm_model_chat.gif\" width=\"600\" alt=\"sample\"\u003e\n\u003c/p\u003e\n\nWith use of OpenVINO™ Test Drive you can:\n+ **Chat with LLMs** and evaluating model performance on your computer or edge device\n+ **Experiment with different text prompts** to generate images using Stable Diffusion and Stable DiffusionXL models\n+ **Transcribe speech from video** using Whisper models, including generation of timestamps\n+ **Run and visualize results of models** trained by Intel® Geti™ using single image inference or batch inference mode\n\n## Installation\n\nDownload the latest release from the [Releases repository](https://storage.openvinotoolkit.org/repositories/openvino_testdrive/).\n\n\u003e [!NOTE]\n\u003e To verify downloaded file integrity, you can generate a SHA-256 of the downloaded file and compare it to the SHA-256 from corresponding `.sha256` file published in Releases repository. \n\n### Installation on Windows\n\n\u003e [!IMPORTANT]\n\u003e For Intel® NPU, please use the Intel® NPU Driver latest available version.\n\n1. Downloading the zip archive [Releases repository](https://storage.openvinotoolkit.org/repositories/openvino_testdrive/) `Windows` folder .\n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/win_inst.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n2. Extract zip archive double-click the MSIX installation package, click `Install` button and it will display the installation process\n\n3. Click on the application name on Windows app list to launch OpenVINO™ Test Drive.\n\n\n## Quick start\n\nUpon starting the application, you can import a model using either Hugging Face for LLMs or upload Intel® Geti™ models from local disk.\n\n### Text generation and LLM performance evaluation\n\n1. Choose a model from predefined set of popular models or pick one from Hugging Face using `Import model` -\u003e `Hugging Face` and import it.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/llm_import.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n2. Pick imported LLM from `My models` section and chat with it using `Playground` tab. You can export LLM via `Export model` button.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/llm_model_chat.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n3. Use `Performance metrics` tab to get LLM performance metrics on your computer.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/metrics.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n### Retrieval-Augmented Generation with LLM\n\n1. It is possible to upload files and create knowledge base for RAG (Retrieval-Augmented Generation) using `Knowledge base` tab\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/rag_base.gif\" width=\"500\"\u003e\n\u003c/p\u003e\nThis knowledge base can be used during text generation with LLM models.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/rag1.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n2. It is also possible to upload document directly using `Playground` tab.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/rag2.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n### Work with Visual Language Models\n\n1. Try Visual Language Model (VLM) for image analysis.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/vlm1.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n2. Pick imported VLM from `My models` section, upload image and analyze it.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/vlm2.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n### Transcribe speech from video\n\n1. Try Whisper for video transcription.\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/st_import.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n2. Pick imported speech-to-text LLM from `My models` section and upload video for transcription. It is also possible to search words in transcript or download it.\n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/video.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n3. Use `Performance metrics` tab to get LLM performance metrics on your computer.\n\n### Image generation\n\n1. Choose an image generation LLM from predefined set of popular models or pick one from Hugging Face using `Import model` -\u003e `Hugging Face` and import it.\n\n2. Pick imported LLM from `My models` section and chat with it to generate image. It is also possible to download generated image.\n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/ig.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n3. Use `Performance metrics` tab to get LLM performance metrics on your computer.\n\nYou can export LLM via `Export model` button.\n\n### Images inference with models trained by Intel® Geti™\n\n1. Download code deployment for the model in OpenVINO format trained by Intel® Geti™. \n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/geti_download.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n\u003e [!NOTE]\n\u003e Please check [Intel® Geti™ documentation](https://docs.geti.intel.com) for more details.\n\n2. Import deployment code into OpenVINO™ Test Drive using `Import model` -\u003e `Local disk` button.\n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/geti_import.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n3. Run and visualize results of inference on individual images using `Live inference` tab.\n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/geti_cv.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n4. For batch inference, use `Batch inference` tab, provide paths to folder with input images in a `Source folder` and specify `Destination folder` for output batch inference results. Click on `Start` to start batch inference.\n\n\u003cp align=\"left\"\u003e\n  \u003cimg src=\"./docs/geti_batch.gif\" width=\"500\"\u003e\n\u003c/p\u003e\n\n## Build\n\nThe application requires the flutter SDK and the dependencies for your specific platform to be installed.\nSecondly, the bindings and its dependencies for your platform to be added to `./bindings`.\n\n1. [Install flutter sdk](https://docs.flutter.dev/get-started/install). Make sure to follow the guide for flutter dependencies.\n2. Build the bindings and put them to `./bindings` folder. OpenVINO™ Test Drive uses bindings to OpenVINO™ GenAI and OpenVINO™ Vision ModelAPI located in `./openvino_bindings` folder. See [readme](./openvino_bindings/README.md) for more details.\n3. Once done you can start the application: `flutter run`\n\n## Ecosystem\n\n- [OpenVINO™](https://github.com/openvinotoolkit/openvino)  - software toolkit for optimizing and deploying deep learning models.\n- [GenAI Repository](https://github.com/openvinotoolkit/openvino.genai) and [OpenVINO Tokenizers](https://github.com/openvinotoolkit/openvino_tokenizers) - resources and tools for developing and optimizing Generative AI applications.\n- [Intel® Geti™](https://docs.geti.intel.com/) - software for building computer vision models.\n- [OpenVINO™ Vision ModelAPI](https://github.com/openvinotoolkit/model_api) - a set of wrapper classes for particular tasks and model architectures, simplifying data preprocess and postprocess as well as routine procedures.\n\n## Contributing\n\nFor those who would like to contribute to the OpenVINO™ Test Drive, please check out [Contribution Guidelines](CONTRIBUTING.md) for more details.\n\n## License\n\nOpenVINO™ Test Drive repository is licensed under [Apache License Version 2.0](LICENSE).\nBy contributing to the project, you agree to the license and copyright terms therein and release your contribution under these terms.\n\nFFmpeg is an open source project licensed under LGPL and GPL. See https://www.ffmpeg.org/legal.html. You are solely responsible for determining if your use of FFmpeg requires any additional licenses. Intel is not responsible for obtaining any such licenses, nor liable for any licensing fees due, in connection with your use of FFmpeg.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopenvinotoolkit%2Fopenvino_testdrive","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fopenvinotoolkit%2Fopenvino_testdrive","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopenvinotoolkit%2Fopenvino_testdrive/lists"}