V-JEPA 2 is a cutting-edge video understanding model developed by the FAIR team under Meta. It extends the pre-training objectives of VJEPA and has industry-leading video understanding capabilities.

Computer VisionTransformers

FairyR1-32B is an efficient large language model developed by Peking University DS-LAB, based on DeepSeek-R1-Distill-Qwen-32B. It achieves a balance between high performance and low-cost inference through an innovative 'distillation-fusion' process.

Natural Language ProcessingTransformersEnglish

FairyR1-32B is a large language model with 32B parameters, developed by PKU-DS-LAB, focusing on text generation tasks.

Natural Language ProcessingGguf

FairyR1-32B is an efficient large language model based on DeepSeek-R1-Distill-Qwen-32B, optimized through distillation and merging processes, excelling in mathematical and programming tasks.

Natural Language ProcessingTransformersEnglish

Meta's PyTorch-based pre-trained language model, compliant with FAIR Non-commercial Research License

Natural Language ProcessingPytorchEnglish

Ettin is the first collection of encoder - only and decoder - only models trained using the same data, architecture, and training method, supporting fair comparison across scales.

Natural Language ProcessingPytorchEnglish

A fast race classification model based on YOLOv8/YOLO11 architecture, trained on the FairFace dataset for facial race classification tasks.

Computer VisionPytorch

Mitsua Likes is a Japanese/English text-to-image latent diffusion model developed based on the concept of collaborative art creation, trained exclusively with explicitly authorized and licensed data, and certified for fair training.

Computer VisionDiffusersMultiple Languages

An image classification model based on Vision Transformer architecture, pre-trained on the ImageNet-21k dataset, suitable for multi-category image classification tasks

Computer VisionTransformers

An image gender classification model based on the Vision Transformer (ViT) architecture with an accuracy of approximately 93.4%

Computer VisionSafetensors

A high-quality Japanese-to-English neural machine translation model developed by MingShiba, optimized based on the fairseq framework and CTranslate2

Natural Language ProcessingFairseqMultiple Languages

This model adopts the FAIR Non-commercial Research License and is suitable for non-commercial research purposes, complying with the FAIR Acceptable Use Policy.

Natural Language ProcessingTransformers

F5-TTS is a flow matching-based voice synthesis model, focusing on fluent and faithful voice synthesis, especially suitable for scenarios like fairy tale narration.

SAM 2.1 Tiny is a lightweight image and video universal segmentation model introduced by Facebook AI Research (FAIR), supporting controllable visual segmentation based on prompts.

Computer VisionCoreml

SAM 2.1 Large is a universal segmentation model released by FAIR, suitable for promptable visual segmentation tasks in images and videos.

Computer VisionCoreml

SAM 2.1 BasePlus is a general-purpose segmentation model released by FAIR in Core ML format, supporting promptable visual segmentation tasks in images and videos.

Computer VisionCoreml

SAM 2 is a foundational model for promptable visual segmentation in images and videos developed by FAIR, supporting efficient segmentation through prompts.

SAM 2 is a foundational model for promptable visual segmentation in images and videos developed by FAIR, supporting efficient segmentation through prompts.

SAM 2 is a foundational model proposed by FAIR to address promptable visual segmentation in images and videos.

SAM 2 is a foundational model for promptable visual segmentation in images and videos developed by FAIR, supporting universal segmentation tasks through prompts.