A unified training engine for multimodal models, supporting vision-language, diffusion, and MoE architectures with FSDP2, Muon, and Liger kernels.
EvolvingLMMs-Lab/NEO
Open-source native vision-language models (NEO series) that process pixels and words in one encoder-free dense architecture, with 2B and 9B releases.
- Звёзды
- 903
- Форки
- 0
- Открытые issues
- 0
- Последний коммит
- 3 месяца назад
Ключевые возможности
- 01Encoder-free dense architecture that unifies pixel and word encoding in one model.
- 022B and 9B model weights released across training stages.
- 03Evaluation guide and code built on VLMEvalKit.
- 04Training code for NEO models in a dedicated guide.
- 05Covers single-image, multi-image, video and spatial understanding tasks.
Сценарии использования
- Researching native or encoder-free multimodal model design.
- Benchmarking vision-language models on image, video and spatial tasks.
- Fine-tuning a compact 2B or 9B VLM on custom image-text data.
Темы
Частые вопросы
- Что такое NEO?
- Open-source native vision-language models (NEO series) that process pixels and words in one encoder-free dense architecture, with 2B and 9B releases.
- NEO можно использовать бесплатно? Какая у него лицензия?
- NEO распространяется с открытым исходным кодом под лицензией Apache-2.0. Перед использованием в коммерческих продуктах проверьте условия лицензии.
- NEO всё ещё поддерживается?
- Обновлялся за последние 6 месяцев. Последний коммит в NEO — 27 июл. 2026 г..
- Для чего можно использовать NEO?
- Researching native or encoder-free multimodal model design. Benchmarking vision-language models on image, video and spatial tasks. Fine-tuning a compact 2B or 9B VLM on custom image-text data.
- Где скачать NEO?
- Исходный код и инструкции по установке — на GitHub: https://github.com/EvolvingLMMs-Lab/NEO
Описание создано ИИ на основе README проекта. Статистика GitHub обновляется ежедневно.
Похожие проекты
A unified multimodal model that handles many computer vision tasks by generating text, images, or both from natural-language instructions.
Open-source native unified multimodal models from SenseNova that handle image understanding, text-to-image generation, and image editing in one architecture.
A 9B-parameter open-weight multimodal model for visual understanding, spatial reasoning, tool use, and embodied-AI research, built on Qwen3.5 and C-RADIOv4-H.
Converts PDFs, Office files, scans, screenshots and diagrams into Markdown by sending pages to your own OpenAI-compatible vision model.
A video LLM trained with streaming speech transcription so it can produce real-time commentary on video, with Gradio and CLI demos.