Search public pages, research tools, and SAVRN solutions.
SAVRN Model Hub
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.
Updated 2026-09-18 · How the library is built
2,760 models, sorted by most downloaded.
AraBART is the first Arabic model in which the encoder and the decoder are pretrained end-to-end, based on BART. AraBART follows the architecture of BART-Base which has 6 encoder and 6 decoder layers and 768 hidden dimensions. In total AraBART has 139M parameters. AraBART achieves the best performance on multiple abstractive summarization datasets, outperforming strong baselines including a pretrained Arabic BERT-based models and multilingual mBART and mT5 models.
The Longformer Encoder-Decoder (LED) for Narrative-Esque Long Text Summarization is a model I fine-tuned from allenai/led-base-16384 to condense extensive technical, academic, and narrative content in a fairly generalizable way. - Ideal for summarizing long narratives, articles, papers, textbooks, and other documents. - the sparknotes-esque style leads to 'explanations' in the summarized content, offering insightful output. The model was trained on the BookSum dataset released by SalesForce, which leads to the bsd-3-clause license. The training process involved 16 epochs with parameters tweaked to facilitate very fine-tuning-type training (super low learning rate). This model is the…
MOSS-Transcribe-Diarize 0.9B is an end-to-end audio understanding model for long-form multi-speaker transcription, diarization, timestamps, and acoustic event awareness. It supports transcription and diarization across 50+ languages, single-pass inference on audio recordings up to 90 minutes long, and custom hotword prompting for domain-specific terms. Given an audio or video file, the model generates a compact speaker-aware transcript in one pass, including timestamps and anonymous speaker labels such as [S01], [S02], and beyond. 2026-07-22: The subtitle Web UI now supports both Simplified Chinese and English. 2026-07-14: MOSS-Transcribe-Diarize won first place in the 2nd MLC-SLM Challenge…
TabFM is a zero-shot tabular foundation model from Google Research. It supports classification and regression on structured/tabular data with mixed numerical and categorical columns, requiring no fine-tuning or hyperparameter search - training examples are passed as context and predictions are made in a single forward pass. This repository contains the JAX/Flax weights stored as Orbax checkpoints. For the PyTorch weights see You can also load directly using the HuggingFace Hub API: Developed by the Google Research team. - Tabular data with numerical and/or categorical columns - Binary and multiclass classification (up to 10 classes) - GPU/TPU acceleration via JAX (use bfloat16 dtype for…
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository. This is V-JEPA 2 ViT-L 256 model with video classification head pretrained on Diving 48 dataset. To run V-JEPA 2 model, ensure you have installed the latest transformers
This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we…
Produced by Reallexi LLC on Reallexi AI Model Builder, a local-first training platform (https://llm.reallexi.io). Hugging Face repository: reallexi/lexi-coder-v4.4. Copyright (c) 2026 Reallexi LLC. All rights reserved.
Model · Video classification
VideoMAE model pre-trained for 2400 epochs in a self-supervised way and fine-tuned in a supervised way on Something-Something V2. It was introduced in the paper VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training by Tong et al. and first released in this repository. Disclaimer: The team releasing VideoMAE did not write a model card for this model so this model card has been written by the Hugging Face team. VideoMAE is an extension of Masked Autoencoders (MAE) to video. The architecture of the model is very similar to that of a standard Vision Transformer (ViT), with a decoder on top for predicting pixel values for masked patches. Videos are…
For the detailed information, see the Overview subcard.
Also, don't confuse APEX-I-MiniPlus (Standard) with a generic baseline APEX-I-Mini. Traditional APEX-I-Mini drops core experts aggressively to 2-bit IQ2S and leaves output.weight at 3-bit Q3KM, which creates a noticeable perplexity hit on complex reasoning tasks. Standard MiniPlus avoids that degradation floor while keeping boundary layers in linear Q3K for single-cycle vectorized AVX2 CPU dequantization (hitting 23 to 26+ tok/s on DDR4 laptops), while protecting output in Q6K and routers in F32. To put the numbers in perspective: this cuts nearly 2 GB off a flat 3-bit quant (approx. 15.6 GB), and weighs only about approx. 1 GB more than a generic APEX-I-Mini (approx. 12.5 GB). For that…
C/C++ — https://github.com/ggml-org/llama.cpp
weighted/imatrix quants of https://huggingface.co/IFM/AmberSafe For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/AmberSafe-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 See…
Architecture: Qwen 3.8 27B (Dense) | Parameters: 27B | Context: 262,144 native / 1,010,000 extensible | Base: Heretic ARA (heretic-org) | Teacher: Claude Fable 5 | Type: Distilled + Abliterated LLM A personal sibling to FoolDev/Janus-35B-HERETIC, which since its 0.9.0 is a Qwen 3.6 35B-A3B MoE — so the two are genuinely different models, not two abliterations of one base (see Janus vs Thanatos). This one is the dense Qwen/Qwen3.8-27B, sharing the teacher (Claude Fable 5) and dataset family, built on heretic-org/Qwen3.8-27B-heretic-ara, an uncensored Heretic ARA-method abliteration, with refusal-trained behavior dialed back at the base layer. (The sibling's -35B- slug names its model…
TabPFNMix classifier is a tabular foundation model that is pre-trained on purely synthetic datasets sampled from a mix of random classifiers. TabPFNMix is based on a 12-layer encoder-decoder Transformer of 37 M parameters. We use a pre-training strategy incorporating in-context learning, similar to that used by TabPFN and TabForestPFN. To use TabPFNMix classifier, install AutoGluon by running: A minimal example showing how to perform fine-tuning and inference using the TabPFNMix classifier: If you find TabPFNMix useful for your research, please consider citing the associated papers: This project is licensed under the Apache-2.0 License.
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we…
This model is a fine-tuned version of on an unknown dataset. The following hyperparameters were used during training: - learningrate: 1e-05 - trainbatchsize: 2 - evalbatchsize: 4 - gradientaccumulationsteps: 4 - totaltrainbatchsize: 8 - lrschedulertype: cosine - lrschedulerwarmupsteps: 0.1 - numepochs: 50 - Transformers 5.12.1 - Pytorch 2.5.1+cu121 - Datasets 5.0.1 - Tokenizers 0.22.2
TimeSformer model pre-trained on Something Something v2. It was introduced in the paper TimeSformer: Is Space-Time Attention All You Need for Video Understanding? by Tong et al. and first released in this repository. Disclaimer: The team releasing TimeSformer did not write a model card for this model so this model card has been written by fcakyon. You can use the raw model for video classification into one of the 174 possible Something Something v2 labels. Here is how to use this model to classify a video: For more code examples, we refer to the documentation.
This model has been pushed to the Hub using the PytorchModelHubMixin integration
This model was finetuned and converted to GGUF format using Unsloth. - unslothQwen3.5-4Bproject-opus-distil1783098194.Q4KM.gguf - unslothQwen3.5-4Bproject-opus-distil1783098194.BF16-mmproj.gguf This was trained 2x faster with Unsloth
Trained models from a comprehensive ML/AI portfolio spanning 10 categories. All models were trained on real Kaggle datasets.
This model is a fine-tuned version of MCG-NJU/videomae-base for binary violence classification (violent / non-violent). It builds on Nikeytas/videomae-crime-detector-production-v1, which was itself fine-tuned from videomae-base on a subset of UCF Crime. Starting from that checkpoint, this model was further fine-tuned on the Bus Violence Dataset to close the domain gap to public-transport surveillance footage. - UCF Crime (jinmang2/ucfcrime) — inherited from the base checkpoint - Bus Violence Dataset (Zenodo) — real moving-bus footage, binary violent / non-violent labels, used for domain-specific fine-tuning Evaluated on a held-out Bus Violence Dataset test split (n = 280). The base…
Model · Text generation
An OpenRLHF GRPO reinforcement-learning checkpoint for Qwen3-4B. - Saved at global step 40 of RL run seededrlbaseramp25stoppengen4kep2ncp10baseq4v3. - This is the best checkpoint by pass@8 so far in this run (evaldefaultpass8 = 0.0421). Trained and validated on the cobalt-train ≤2/64 frontier (canonical cleaneval prompts): 1833 train / 112 held-out val problems the base model solved on at most 2 of 64 samples under the iidcanonical@64 hardness scan. Val evals sample at temperature 1.0 (matching the cleaneval frontier eval). Reward signal: binary code-correctness (1.0 if the generated program passes the problem's tests, otherwise 0.0). Eval metrics at this checkpoint (held-out val, 8…
Hand-picked starting points, each with the reason it exists.
Collection · 4 entries
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 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
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
Recognition and synthesis models, grouped so the two directions are easy to compare.
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.
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.
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.
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.
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
The same open-weight model priced by every host that serves it, per million tokens.
Research Hub
Moratoriums, permits, power, water and capital behind the facilities that run these models.
Method
Sources, evidence labels, refresh behaviour, and the limits of every comparison here.
