Hi, I'm a Machine Learning Engineer, Hugging Face Fellow ML 🤗, Computer Vision Enthusiast.

• Qwen-Image-Edit-2511-LoRAs-Fast: Collection demo of QIE-2511 LoRAs, 1.5K+ likes, 3.5M+ runs. [Collection]
• FireRed-Image-Edit-1.0-Fast: Fast 4-step inference demo of the FireRed Image Edit model, 1.4K+ likes, 2.7M+ runs. [Collection]
• Qwen-Image-Edit-2509-LoRAs-Fast: Collection demo of QIE-2509 LoRAs, 760+ likes, 2.2M+ runs. [Collection]
• FLUX.2-Klein-LoRA-Studio: Collection demo of Klein LoRAs, 120+ likes, 145K+ runs. [Collection]
• FLUX-LoRA-DLC: FLUX.1-dev diffusion model with 255+ community LoRAs, 1.09K+ likes, 70K+ runs. [Collection]
• Multimodal-OCR: OCR for images and videos using state-of-the-art vision-language models, 40K+ runs, 90K+ visits. [Collection]
• Multimodal-VLM-Thinking: VLMs for captioning, OCR, reasoning, and multimodal tasks, 2.06K+ runs, 11.2K+ visits. [Collection]
• Qwen3-VL-Outpost: VLM for image and video understanding with multilingual support, 6.2K+ runs, 49.9K+ visits. [Collection]
• Flux Realism: Hyper-realistic image generation with FLUX.1-dev and Super Realism LoRA, 39.5K+ runs, 127.7K+ visits. [Collection]

• OneDecision-VisionGuard-27B-SFT: OneDecision-VisionGuard-27B-SFT is a dense 27B multimodal model based on Qwen/Qwen3.8-27B, designed to classify images as Safe or NSFW.[Collection]
• Gliese-OCR-7B-Post1.0: Enhanced document retrieval, content extraction, and analysis, built on Camel-Doc-OCR-062825.[Collection]
• DeepCaption-VLA-7B: Generates precise, descriptive image captions highlighting visual properties, object attributes. [Collection]
• Camel-Doc-OCR: Document retrieval, content extraction, and analysis. (v2 080125) [Collection]
• SigLIP2-0.1B-DownStream: Domain-specific image classification models fine-tuned from siglip2 for multi-label tasks. [Base]
• Lumian2-VLR-7B: VLM for fine-grained multimodal reasoning, image/video captioning, and document comprehension with explainable step-by-step reasoning. [Demo]

• Galactic-Qwen-14B: Top mid-range 14B model, ranked 59th, overall score 43.56. [Leaderboard]
• Gauss-Opus-14B: Strong in math, ranked 356th, MATH Level 5 score 57.55. [Leaderboard]
• Sombrero-Opus-14B: All-rounder mid-range 14B, ranked 104th, score 42.32. [Leaderboard]
• Dinobot-Opus-14B: IFEval score 82.40, ranked 132nd, overall 41.77. [Leaderboard]
• Qwen2-VL-OCR-2B: Edge-device VLM for handwriting, LaTeX, bills, and receipts, 250k+ downloads. [Run Demo]

• Stranger Vision: Community for model modification and experimentation, < 1K downloads. [Collection]
• Stranger Zone: Illustration adapters for diffusion models, 2M+ downloads. [Collection]
• Stranger Guard: Image safety-guard models, 10k+ downloads. [Collection]
• Stranger Operations: Model Fostering, Operations, and Cycle
• Stranger Tools: Tools, Wheels, Fun