Open-Weight Models

Model · Time series forecasting

granite-timeseries-ttm-r2

IBM Granite

TinyTimeMixers (TTMs) are compact pre-trained models for Multivariate Time-Series Forecasting, open-sourced by IBM Research. With model sizes starting from 1M params, TTM introduces the notion of the first-ever “tiny” pre-trained models for Time-Series Forecasting. The paper describing TTM was accepted at NeurIPS 24. TTM outperforms other models demanding billions of parameters in several popular zero-shot and few-shot forecasting benchmarks. TTMs are lightweight forecasters, pre-trained on publicly available time series data with various augmentations. TTM provides state-of-the-art zero-shot forecasts and can easily be fine-tuned for multi-variate forecasts with just 5% of the training…

Model · Image segmentation

segformer-b2-finetuned-ade-512-512

NVIDIA

SegFormer model fine-tuned on ADE20k at resolution 512x512. It was introduced in the paper SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers by Xie et al. and first released in this repository. Disclaimer: The team releasing SegFormer did not write a model card for this model so this model card has been written by the Hugging Face team. SegFormer consists of a hierarchical Transformer encoder and a lightweight all-MLP decode head to achieve great results on semantic segmentation benchmarks such as ADE20K and Cityscapes. The hierarchical Transformer is first pre-trained on ImageNet-1k, after which a decode head is added and fine-tuned altogether on a…

This repository contains the official trained policies for The files are organized by task environment and training seed: The play.py script in the JoSE repository downloads a pretrained model and executes it in the corresponding environment, allowing agent behaviour to be visualized. The script is called from the repo root as follows: Seeds vary by hand and task. See the HuggingFace model directory to see which seeds are available for each task-hand combination.

This model is based on DistilBERT and has been fine-tuned for multilabel classification of Emails and URLs as safe or potentially phishing. - Base Architecture: DistilBERT - Task: Multilabel Classification - Fine-tuning Framework: Hugging Face Trainer API - Training Duration: 3 epochs - Accuracy: 99.58 - F1-score: 99.579 - Precision: 99.583 - Recall: 99.58 The model was trained on a custom dataset of Emails and URLs labeled as legitimate or phishing. The dataset is available at cybersectony/PhishingEmailDetectionv2.0 on the Hugging Face Hub.

Model · Text to speech

higgs-tts-2-3b-base

Boson AI

Check our open-source repository https://github.com/boson-ai/higgs-audio for more details! We are open-sourcing Higgs TTS 2, a powerful audio foundation model pretrained on over 10 million hours of audio data and a diverse set of text data. Despite having no post-training or fine-tuning, Higgs TTS 2 excels in expressive audio generation, thanks to its deep language and acoustic understanding. On EmergentTTS-Eval, the model achieves win rates of 75.7% and 55.7% over "gpt-4o-mini-tts" on the "Emotions" and "Questions" categories, respectively. It also obtains state-of-the-art performance on traditional TTS benchmarks like Seed-TTS Eval and Emotional Speech Dataset (ESD). Moreover, the model…

Model · Text to image

playground-v2.5-1024px-aesthetic

Playground

This repository contains a model that generates highly aesthetic images of resolution 1024x1024, as well as portrait and landscape aspect ratios. You can use the model with Hugging Face Diffusers. Playground v2.5 is a diffusion-based text-to-image generative model, and a successor to Playground v2. Playground v2.5 is the state-of-the-art open-source model in aesthetic quality. Our user studies demonstrate that our model outperforms SDXL, Playground v2, PixArt-α, DALL-E 3, and Midjourney 5.2. For details on the development and training of our model, please refer to our blog post and technical report. Install diffusers >= 0.27.0 and the relevant dependencies. - The pipeline uses the…

Model · Any to any

gemma-4-E4B-it-uncensored-heretic

LLMfan46

Creating these models takes significant time, work and compute. If you find them useful consider supporting me: Your help will motivate me and would go into further improving my workflow and coverings fees for storage, compute and may even help uncensoring bigger model with rental Cloud GPUs. attn.oproj Lower refusals indicate fewer content restrictions, while lower KL divergence indicates more closeness to the original model's baseline. Higher refusals cause more rejections, objections, pushbacks, lecturing, censorship, softening and deflections. PIQA (Physical Intuition Question Answering) a ~1,800 questions tests common-sense understanding of how the physical world works with benchmark…

Model · Speech recognition

seamless-m4t-v2-large

AI at Meta

SeamlessM4T is our foundational all-in-one Massively Multilingual and Multimodal Machine Translation model delivering high-quality translation for speech and text in nearly 100 languages. SeamlessM4T models support the tasks of: - Automatic speech recognition (ASR). - 101 languages for speech input. - 96 Languages for text input/output. - 35 languages for speech output. We are releasing SeamlessM4T v2, an updated version with our novel UnitY2 architecture. This new model improves over SeamlessM4T v1 in quality as well as inference speed in speech generation tasks. The v2 version of SeamlessM4T is a multitask adaptation of our novel UnitY2 architecture. Unity2 with its hierarchical…

Model · Token classification

llmlingua-2-bert-base-multilingual-cased-meetingbank

Microsoft

This model was introduced in the paper LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression (Pan et al, 2024). It is a BERT multilingual base model (cased) finetuned to perform token classification for task agnostic prompt compression. The probability $p{preserve}$ of each token $xi$ is used as the metric for compression. This model is trained on the extractive text compression dataset constructed with the methodology proposed in the LLMLingua-2, using training examples from MeetingBank (Hu et al, 2023) as the seed data. You can evaluate the model on downstream tasks such as question answering (QA) and summarization over compressed meeting transcripts…

Model · Token classification

xlm-roberta-base-ner-hrl

David Adelani

Hugging Face's logo - multilingual xlm-roberta-base-ner-hrl is a Named Entity Recognition model for 10 high resourced languages (Arabic, German, English, Spanish, French, Italian, Latvian, Dutch, Portuguese and Chinese) based on a fine-tuned XLM-RoBERTa base model. It has been trained to recognize three types of entities: location (LOC), organizations (ORG), and person (PER). Specifically, this model is a xlm-roberta-base model that was fine-tuned on an aggregation of 10 high-resourced languages You can use this model with Transformers pipeline for NER. This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well…

Model · Translation

sanskrit-en-custom-transformer

Praveen Kumar

A compact encoder–decoder Transformer trained from scratch on 10,000 Sanskrit–English pairs (NLU Assignment 2), with a joint SentencePiece BPE vocabulary. It is small and fast — meant to be efficient rather than to match large pretrained models. This is not a Transformers architecture, so it ships with a self-contained modeling.py. Files: pytorchmodel.bin (weights), spm.model (tokenizer), config.json (hyperparameters), modeling.py (model + load/translate helpers). pip install gradio first. To host it, create a Hugging Face Space (SDK: Gradio) with an app.py (the load + respond code) and a requirements.txt of torch sentencepiece huggingfacehub gradio. Trained only on the provided dataset…

Model · Image to image

FLUX.2-klein-base-4B

Black Forest Labs

The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in as low as under a second. Built for applications that require real-time image generation without sacrificing quality, and runs on consumer hardware, with as little as 13GB VRAM. FLUX.2 [klein] 4B Base is a 4 billion parameter rectified flow transformer capable of generating images from text descriptions and supports multi-reference editing capabilities. It's a full-capacity foundation model. Undistilled, preserving complete training signal for maximum flexibility. Ideal for…

detr-doc-table-detection is a model trained to detect both Bordered and Borderless tables in documents, based on facebook/detr-resnet-50. This model can be used for the task of object detection. The model should not be used to intentionally create hostile or alienating environments for people. Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups. Users (both direct and downstream) should be made aware of the risks, biases and…

Model · Image segmentation

mask2former-swin-large-ade-semantic

AI at Meta

Mask2Former model trained on ADE20k semantic segmentation (large-sized version, Swin backbone). It was introduced in the paper Masked-attention Mask Transformer for Universal Image Segmentation and first released in this repository. Disclaimer: The team releasing Mask2Former did not write a model card for this model so this model card has been written by the Hugging Face team. Mask2Former addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation. Mask2Former outperforms the previous SOTA, MaskFormer both in terms of performance an efficiency by (i)…

Model · Translation

nllb-200-distilled-1.3B

AI at Meta

This is the model card of NLLB-200's distilled 1.3B variant. Here are the metrics for that particular checkpoint. - Information about training algorithms, parameters, fairness constraints or other applied approaches, and features. The exact training algorithm, data and the strategies to handle data imbalances for high and low resource languages that were used to train NLLB-200 is described in the paper. - Paper or other resource for more information NLLB Team et al, No Language Left Behind: Scaling Human-Centered Machine Translation, Arxiv, 2022 - Where to send questions or comments about the model: https://github.com/facebookresearch/fairseq/issues • Model performance measures: NLLB-200…

Model · Text to speech

tts-hifigan-libritts-22050Hz

SpeechBrain

This repository provides all the necessary tools for using a HiFIGAN vocoder trained with LibriTTS (with multiple speakers). The sample rate used for the vocoder is 22050 Hz. The pre-trained model takes in input a spectrogram and produces a waveform in output. Typically, a vocoder is used after a TTS model that converts an input text into a spectrogram. Alternatives to this models are the following: - tts-hifigan-libritts-16kHz (same model trained on the same dataset, but for a sample rate of 16000 Hz) - tts-hifigan-ljspeech (same model trained on LJSpeech for a sample rate of 22050 Hz). Please notice that we encourage you to read our tutorials and learn more about To perform inference on…

Model · Image to image

FLUX.2-klein-4B-GGUF

Unsloth AI

This is a GGUF quantized version of FLUX.2-klein-4B. unsloth/FLUX.2-klein-4B-GGUF uses Unsloth Dynamic 2.0 methodology for SOTA performance. - Important layers are upcasted to higher precision. - Uses tooling from ComfyUI-GGUF by city96. The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in as low as under a second. Built for applications that require real-time image generation without sacrificing quality, and runs on consumer hardware, with as little as 13GB VRAM. FLUX.2 [klein] 4B is a 4 billion parameter rectified flow…

Model · Text to image

Z-Image-Turbo-GGUF

Unsloth AI

Welcome to the official repository for the Z-Image（造相）project! Z-Image is a powerful and highly efficient image generation model with 6B parameters. Currently there are three variants: - Z-Image-Turbo – A distilled version of Z-Image that matches or exceeds leading competitors with only 8 NFEs (Number of Function Evaluations). It offers sub-second inference latency on enterprise-grade H800 GPUs and fits comfortably within 16G VRAM consumer devices. It excels in photorealistic image generation, bilingual text rendering (English & Chinese), and robust instruction adherence. - Z-Image-Base – The non-distilled foundation model. By releasing this checkpoint, we aim to unlock the full potential…

Model · Text to video

Wan2.1-T2V-1.3B-Diffusers

Wan-AI

In this repository, we present Wan2.1, a comprehensive and open suite of video foundation models that pushes the boundaries of video generation. Wan2.1 offers these key features: This repository hosts our T2V-1.3B model, a versatile solution for video generation that is compatible with nearly all consumer-grade GPUs. In this way, we hope that Wan2.1 can serve as an easy-to-use tool for more creative teams in video creation, providing a high-quality foundational model for academic teams with limited computing resources. This will facilitate both the rapid development of the video creation community and the swift advancement of video technology. Your browser does not support the video tag.…

Model · Fill mask

caduceus-ps_seqlen-131k_d_model-256_n_layer-16

Kuleshov Group

To use the pre-trained model for masked language modeling, use the following snippet: Alternatively, you can instantiate a model from scratch to train on your own data as follows: This is the Caduceus-PS model with hidden dimension 256 and 16 MambaDNA layers. This model is reverse complement (RC) equivariant and thus no RC data augmentation is required when training this model, either during pre-training or for downstream fine-tuning. Note that the model hidden state will be twice that of a non-RC equivariant counterpart. For downstream task training and inference, and to ensure RC invariant outputs at downstream time, one can either run the downstream model on the hidden state and its RC…

Model · Image to video

LTX-2.3-GGUF

Unsloth AI

This is a GGUF quantized version of LTX-2.3. unsloth/LTX-2.3-GGUF uses Unsloth Dynamic 2.0 methodology for SOTA performance. - Important layers are upcasted to higher precision. - Uses tooling from ComfyUI-GGUF by city96. There are two sets of GGUF's published. One for the dev model and one for the distilled. The distilled model is optimized for few step generation, think 4-8 steps. dev on the other hand needs more steps at least 20, but you get better outputs. The distilled variant is useful as a drafting model or a refining model. In fact the workflow published below, uses the distilled lora on top of the dev model to refine the intial output. Download the mp4 in the repo and open it with…

Model · Text to speech

Qwen3-TTS-12Hz-1.7B-VoiceDesign

Qwen

We release Qwen3-TTS, a series of powerful speech generation models developed by Qwen, offering comprehensive support for voice cloning, voice design, ultra-high-quality human-like speech generation, and natural language-based voice control. Qwen3-TTS covers 10 major languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian) as well as multiple dialectal voice profiles. Key features: Install the qwen-tts Python package from PyPI: Zero-shot speech generation on the Seed-TTS test set (Word Error Rate (WER, ↓)): If you find our paper and code useful in your research, please consider giving a star and citation

Model · Tabular classification

mitra-classifier

Autogluon

Mitra classifier is a tabular foundation model that is pre-trained on purely synthetic datasets sampled from a mix of random classifiers. Mitra is based on a 12-layer Transformer of 72 M parameters, pre-trained by incorporating an in-context learning paradigm. To use Mitra classifier, install AutoGluon by running: A minimal example showing how to perform inference using the Mitra classifier: A minimal example showing how to perform fine-tuning using the Mitra classifier: This project is licensed under the Apache-2.0 License. Amazon Science blog: Mitra: Mixed synthetic priors for enhancing tabular foundation models

Model · Fill mask

deberta-base

Microsoft

DeBERTa improves the BERT and RoBERTa models using disentangled attention and enhanced mask decoder. It outperforms BERT and RoBERTa on majority of NLU tasks with 80GB training data. Please check the official repository for more details and updates. We present the dev results on SQuAD 1.1/2.0 and MNLI tasks. If you find DeBERTa useful for your work, please cite the following paper