LW-DETR, a Light-Weight DEtection TRansformer model, is designed to be a real-time object detection alternative that outperforms conventional convolutional (YOLO-style) and earlier transformer-based (DETR) methods in terms of speed and accuracy trade-off. It was introduced in the paper LW-DETR: A Transformer Replacement to YOLO for Real-Time Detection by Chen et al. and first released in this repository. Disclaimer: This model was originally contributed by stevenbucaille in transformers. LW-DETR is an end-to-end object detection model that uses a Vision Transformer (ViT) backbone as its encoder, a simple convolutional projector, and a shallow DETR decoder. The core philosophy is to leverage…
SAVRN Model Hub
Open-Weight Models
An open-weight model is an AI model whose trained weights are published for anyone to download. The weights are what the model learned in training. With a copy of them you can run the model on hardware you control and train it further on your own data.
Open weights are not the same as open source. Many publishers release the weights without the training data or code, and the license sets what you may do with the model. This library puts each model's full card, architecture, files, license and published evaluations on one page.
2,760Models
859Datasets
254Papers
1,692Publishers
5,040Sourced relationships
Updated 2026-09-18 · How the library is built
2,760 models, sorted by most downloaded.
Model obtained by Fine Tuning 'facebook/bart-large-xsum' using AMI Meeting Corpus, SAMSUM Dataset, DIALOGSUM Dataset, XSUM Dataset!
Update Feb 14, 2025: Chronos-Bolt & original Chronos models are now available on Amazon SageMaker JumpStart! Check out the tutorial notebook to learn how to deploy Chronos endpoints for production use in a few lines of code. Update Nov 27, 2024: We have released Chronos-Bolt models that are more accurate (5% lower error), up to 250 times faster and 20 times more memory-efficient than the original Chronos models of the same size. Check out the new models here. Chronos is a family of pretrained time series forecasting models based on language model architectures. A time series is transformed into a sequence of tokens via scaling and quantization, and a language model is trained on these…
t0-alpha is an open-weights time-series forecasting foundation model from The Forecasting Company. t0 is a transformer-based model that produces probabilistic multi-horizon forecasts and natively operates on multiple covariates. t0-alpha is the first public iteration of the model. You can use t0 on Retrocast, The Forecasting Company's platform for forecasting on your own data and comparing forecasts across open-weight models. Model family: t0-alpha (PyTorch/MLX) · French national electricity demand in Retrocast. Data: Enedis open data. t0-alpha is an alpha release intended for research, experimentation, and applied forecasting evaluation. t0-alpha is intended for probabilistic time-series…
Pretrained model on English language using a masked language modeling (MLM) objective. It was introduced in this paper and first released in this repository. This model is uncased: it does not make a difference between english and English. Differently to other BERT models, this model was trained with a new technique: Whole Word Masking. In this case, all of the tokens corresponding to a word are masked at once. The overall masking rate remains the same. The training is identical -- each masked WordPiece token is predicted independently. After pre-training, this model was fine-tuned on the SQuAD dataset with one of our fine-tuning scripts. See below for more information regarding this…
This model ships a Multi-Token Prediction drafter at the repo root (mtp-gemma-4-E2B-it.gguf, a near-lossless smart Q40). A recent llama.cpp auto-discovers it from -hf, so you do not pass --model-draft: The drafter shares the target's KV cache and does not change the output (the target verifies every drafted token). See the MTP/ folder for the other precisions and explicit usage. Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context…
PatchTST-FM-r2, a state-of-the-art zero-shot time series foundation model, represents a continuation of the well-recognized PatchTST model series, building on the original PatchTST and its zero-shot variant PatchTST-FM-r1. PatchTST-FM-r2 brings architectural enhancements as well as an expanded training base on top of its predecessor PatchTST-FM-r1. As of August 31, 2026 Granite-TimeSeries-PatchTST-FM-r2 is the top performing zero-shot model released under a permissive, commercial-friendly open-source license on the GIFT-Eval benchmark. Granite-TimeSeries-PatchTST-FM-r2 ranks #2 when considering all zero-shot, replicable models (see below for more details). The architectural changes in r2…
The DistilBERT model was proposed in the blog post Smaller, faster, cheaper, lighter: Introducing DistilBERT, adistilled version of BERT, and the paper DistilBERT, adistilled version of BERT: smaller, faster, cheaper and lighter. DistilBERT is a small, fast, cheap and light Transformer model trained by distilling BERT base. It has 40% less parameters than bert-base-uncased, runs 60% faster while preserving over 95% of BERT's performances as measured on the GLUE language understanding benchmark. This model is a fine-tune checkpoint of DistilBERT-base-uncased, fine-tuned using (a second step of) knowledge distillation on SQuAD v1.1. - See this repository for more about Distil\ (a class of…
PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training We introduce PaddleOCR-VL-1.6, an upgraded compact document parsing model built upon PaddleOCR-VL-1.5. PaddleOCR-VL-1.6 introduces a region-aware data optimization framework that identifies weak regions from the previous model, applies targeted enhancement to those regions, and improves the reliability of supervision signals. It further adopts a progressive post-training recipe based on curated data selection and reinforcement learning, pushing model performance to a higher level through staged optimization. PaddleOCR-VL-1.6 achieves a new state-of-the-art score…
mms - vits pipelinetag: text-to-speech This repository contains the Hindi (hin) language text-to-speech (TTS) model checkpoint. This model is part of Facebook's Massively Multilingual Speech project, aiming to provide speech technology across a diverse range of languages. You can find more details about the supported languages and their ISO 639-3 codes in the MMS Language Coverage Overview, and see all MMS-TTS checkpoints on the Hugging Face Hub: facebook/mms-tts. MMS-TTS is available in the Transformers library from version 4.33 onwards. VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) is an end-to-end speech synthesis model that predicts a speech…
Update Feb 14, 2025: Chronos-Bolt & original Chronos models are now available on Amazon SageMaker JumpStart! Check out the tutorial notebook to learn how to deploy Chronos endpoints for production use in a few lines of code. Update Nov 27, 2024: We have released Chronos-Bolt models that are more accurate (5% lower error), up to 250 times faster and 20 times more memory-efficient than the original Chronos models of the same size. Check out the new models here. Chronos is a family of pretrained time series forecasting models based on language model architectures. A time series is transformed into a sequence of tokens via scaling and quantization, and a language model is trained on these…
Pretrained model on English language using a masked language modeling (MLM) objective. It was introduced in this paper and first released in this repository. This model is cased: it makes a difference between english and English. Differently to other BERT models, this model was trained with a new technique: Whole Word Masking. In this case, all of the tokens corresponding to a word are masked at once. The overall masking rate remains the same. The training is identical -- each masked WordPiece token is predicted independently. After pre-training, this model was fine-tuned on the SQuAD dataset with one of our fine-tuning scripts. See below for more information regarding this fine-tuning.…
source languages: nl; target languages: fr; OPUS readme: nl-fr; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
GGUF files released for the audio.cpp echotts model. Includes the s1 codec weights. f16 and q80 This work is not affiliated or endorsed by the original author, Jordan Darefsky (https://huggingface.co/jordand/echo-tts-base) Echo-TTS is a diffusion-based text-to-speech model. The weights in this repository are intended for research and non-commercial use only. By using this model, you agree not to use it for: - Deception, fraud, or impersonation, including: - Generating audio that mimics a real person’s voice without their explicit consent. - Creating deepfakes meant to mislead others about who is speaking. - Harassment, abuse, or discrimination, including targeted abuse or hate content…
Mask2Former model trained on COCO instance 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)…
The comfy-native cuts of the MiniMax-H3 × Z-Image graft: Z-Image's spatial-attention profile on H3's engine — richer sets and textures, same identity, no per-shot sharpening creep. Full story, demos and verification on the GGUF page. Load with the plain Load Diffusion Model node, ComfyUI 0.32+. Files are the pruned H3 builds with the graft baked in (zs05 = late-block gains, dose 0.5): - bf16 — the master (ref2va) - comfy-fp8 / fp8e5m2 — fp8 scaled - comfy-int8 / int8convrot — the fast pick on RTX 50 - comfy-w4a8 / w4a4 / nvfp4 — 4-bit family for 16 GB cards (w4a8 is the quality pick; nvfp4 is Blackwell-native, emulated elsewhere) - comfy-mxfp8 — 8-bit microscaling, Blackwell-specialized…
latinPP-OCRv5mobilerec is one of the PP-OCRv5rec that are the latest generation text line recognition models developed by PaddleOCR team. It aims to efficiently and accurately support the recognition of Korean. The key accuracy metrics are as follow: Note: If any character (including punctuation) in a line was incorrect, the entire line was marked as wrong. This ensures higher accuracy in practical applications. Please refer to the following commands to install PaddlePaddle using pip: For details about PaddlePaddle installation, please refer to the PaddlePaddle official website. Install the latest version of the PaddleOCR inference package from PyPI: You can quickly experience the…
RF-DETR is a real-time detection transformer family introduced in RF-DETR: Neural Architecture Search for Real-Time Detection Transformers by Robinson et al. and integrated in Transformers via PR #36895. RF-DETR is an end-to-end instance segmentation model that combines ideas from LW-DETR and Deformable DETR: a DINOv2-with-registers style ViT backbone (with an RF-DETR windowing pattern for efficient attention), a multi-scale projector between encoder and decoder, and a multi-scale deformable DETR decoder extended with an instance-segmentation head. You can use the raw model for instance segmentation; it predicts per-instance masks together with bounding boxes and class scores. See the model…
source languages: fr; target languages: de; OPUS readme: fr-de; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
This is a modified version of google/translategemma-12b-it optimized for deployment with vLLM. No retraining was performed. Only configuration files and the chat template were modified. Model weights are identical to the original. As of 2025-01-29, vLLM does not natively support TranslateGemma's custom structured input format. See vllm-project/vllm#32446 for the upstream tracking issue. Until that is merged, this repo provides a workaround by modifying configuration files to make TranslateGemma compatible with vLLM's standard chat API. This conversion is based entirely on the work done by Infomaniak-AI/vllm-translategemma-4b-it. The same conversion approach was applied to the 12B model.…
Mask2Former model trained on Mapillary Vistas 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…
source languages: sv; target languages: en; OPUS readme: sv-en; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
BEiT model pre-trained in a self-supervised fashion on ImageNet-21k (14 million images, 21,841 classes) at resolution 224x224, and fine-tuned on ADE20k (an important benchmark for semantic segmentation of images) at resolution 640x640. It was introduced in the paper BEIT: BERT Pre-Training of Image Transformers by Hangbo Bao, Li Dong and Furu Wei and first released in this repository. Disclaimer: The team releasing BEiT did not write a model card for this model so this model card has been written by the Hugging Face team. The BEiT model is a Vision Transformer (ViT), which is a transformer encoder model (BERT-like). In contrast to the original ViT model, BEiT is pretrained on a large…
TrOCR model fine-tuned on the SROIE dataset. It was introduced in the paper TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models by Li et al. and first released in this repository. The TrOCR model is an encoder-decoder model, consisting of an image Transformer as encoder, and a text Transformer as decoder. The image encoder was initialized from the weights of DeiT, while the text decoder was initialized from the weights of UniLM. Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder. Next…
Model Collections
Hand-picked starting points, each with the reason it exists.
Collection · 4 entries
Embedding models for retrieval
Sentence and document embedding models used to build retrieval systems. Dimension and sequence length matter more than size here, and both come from the publisher.
Collection · 4 entries
Models that fit on one accelerator
Models whose publisher-reported parameter count puts them within reach of a single accelerator at common precisions. Memory needed depends on precision and serving configuration, so treat the parameter count as the starting point, not the answer.
Collection · 6 entries
Open-weight text models worth knowing
Widely used open-weight language models, chosen because each one is a distinct family rather than a variant of the one above it. Selection, not a ranking.
Collection · 3 entries
Speech and audio models
Recognition and synthesis models, grouped so the two directions are easy to compare.
Open-Weight Models Explained
What is an open-weight model?
An AI model whose trained weights are published for anyone to download, so it can be run, tested and fine-tuned on hardware the user controls.
Is an open-weight model the same as open source?
Not always. Open weights means the trained model can be downloaded. Open source usually also means the training code and data are available and the license allows broad reuse. Many open-weight models release the weights only.
Can I use an open-weight model commercially?
It depends on the license. Apache 2.0 and MIT allow commercial use. Other licenses limit it, for example to non-commercial use or below a set number of users. Every model page here shows its license.
How much memory does an open-weight model need?
About two bytes per parameter at 16-bit precision, so a 7-billion-parameter model needs roughly 14 GB for its weights, plus memory for the context it processes. Each model page lists its parameter count and the size of its files.
Related SAVRN Research
The hub sits beside SAVRN's market data and infrastructure research: what models cost to run, and what it takes to run them.
SAVRN Index
What open models cost to run
The same open-weight model priced by every host that serves it, per million tokens.
Research Hub
Data center trackers and maps
Moratoriums, permits, power, water and capital behind the facilities that run these models.
Method
How the Model Hub is built
Sources, evidence labels, refresh behaviour, and the limits of every comparison here.