# Open-Weight Models Under Creative Commons Attribution 4.0 Source: https://savrn.com/models/licenses/cc-by-4-0 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- SAVRN Model Hub · Models by License # Open-Weight Models Under Creative Commons Attribution 4.0 172 open-weight models released under Creative Commons Attribution 4.0 in the SAVRN Model Hub, with Deepset, Helsinki-NLP Research Group and David Larrea publishing the most. 172Models 114Publishers 17K to 3BParameter range 1Licenses YesCommercial use ## What Creative Commons Attribution 4.0 Allows CC BY 4.0 permits sharing and adapting the work, including commercially, provided the creator is credited and changes are indicated. [Read the license text](https://creativecommons.org/licenses/by/4.0/). ## SAVRN's Take Credit the creator, mark your changes, and the rest is yours. Those are the only conditions the license sets for the 54 models in our catalog under Creative Commons Attribution 4.0, and commercial use is permitted outright. For an operator that means a fine-tuned copy can sit behind a paid product, and the two obligations that travel with it, the credit line and the notice of what changed, belong in the service documentation before the first customer signs, not after. The catalog under this license leans toward speech and language tooling. Deepset publishes 12 of the 54, the Helsinki-NLP Research Group 8, Handy 4, Scvi Tools 3, and Pyannote, MLX Community, Allegro ML Research and ESPnet 2 each. Question answering accounts for 12 models, translation 10, speech recognition 8, text to speech 3 and image classification 2. Downloads are led by Pyannote: wespeaker-voxceleb-resnet34-LM at 6,527,409 a month and speaker-diarization-community-1 at 5,203,396, then Cardiff NLP's twitter-roberta-base-sentiment-latest at 3,141,717. The parakeet speech recognition ports from MLX Community and Handy each clear a million. The parakeet models are small. The 627M parakeet-tdt-0.6b-v3 needs 1.5 GB at 16-bit and 0.4 GB at 4-bit, NVIDIA's 1.1B parakeet-ctc-1.1b needs 2.6 GB at 16-bit, and the cheapest host on the Index for either is one MI300X at $1.85 an hour. Check the access flag before committing: speaker-diarization-community-1 is gated while the rest of the top eight are open access. SAVRN Research, 2026-09-18 ## Most Downloaded | Model | Publisher | Parameters | License | Monthly downloads | Cheapest GPUs at 16-bit | | --- | --- | --- | --- | --- | --- | | [wespeaker-voxceleb-resnet34-LM](https://savrn.com/models/wespeaker-voxceleb-resnet34-lm) | Pyannote | — | cc-by-4.0 | 5.8M | — | | [speaker-diarization-community-1](https://savrn.com/models/speaker-diarization-community-1) | Pyannote | — | cc-by-4.0 | 5M | — | | [parakeet-tdt-0.6b-v3](https://savrn.com/models/parakeet-tdt-0-6b-v3) | MLX Community | 627M | cc-by-4.0 | 2.5M | 1x MI300X, $1.85/hr | | [twitter-roberta-base-sentiment-latest](https://savrn.com/models/twitter-roberta-base-sentiment-latest) | Cardiff NLP | — | cc-by-4.0 | 2.4M | — | | [wav2vec2-xls-r-300m-cv7-turkish](https://savrn.com/models/wav2vec2-xls-r-300m-cv7-turkish) | Muhammet Poyraz | — | cc-by-4.0 | 1.1M | — | | [parakeet-tdt-0.6b-v2](https://savrn.com/models/parakeet-tdt-0-6b-v2) | MLX Community | 618M | cc-by-4.0 | 1.1M | 1x MI300X, $1.85/hr | | [parakeet-ctc-1.1b](https://savrn.com/models/parakeet-ctc-1-1b) | NVIDIA | 1.1B | cc-by-4.0 | 796.2k | 1x MI300X, $1.85/hr | | [audiobox-aesthetics](https://savrn.com/models/audiobox-aesthetics) | AI at Meta | 104M | cc-by-4.0 | 635.2k | 1x MI300X, $1.85/hr | | [roberta-base-squad2](https://savrn.com/models/roberta-base-squad2) | Deepset | 124M | cc-by-4.0 | 559.7k | 1x MI300X, $1.85/hr | | [opus-mt-en-de](https://savrn.com/models/opus-mt-en-de) | Helsinki-NLP Research Group | — | cc-by-4.0 | 442.9k | — | ## Who Publishes Them | Publisher | Models | | --- | --- | | [Deepset](https://savrn.com/model-publishers/deepset) | 12 | | [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) | 8 | | [David Larrea](https://savrn.com/model-publishers/davidalarrea) | 8 | | [Dimitar Gueorguiev](https://savrn.com/model-publishers/dimitarpg13) | 7 | | [Josh Leverette](https://savrn.com/model-publishers/coder543) | 6 | | [Nish](https://savrn.com/model-publishers/nishparadox) | 4 | ## All 172 Models, Page 1 of 3 Model ### [wespeaker-voxceleb-resnet34-LM](https://savrn.com/models/wespeaker-voxceleb-resnet34-lm) [Pyannote](https://savrn.com/model-publishers/pyannote) Using this open-source model in production? Consider switching to pyannoteAI for better and faster options. This model requires pyannote.audio version 3.1 or higher. This is a wrapper around WeSpeaker wespeaker-voxceleb-resnet34-LM pretrained speaker embedding model, for use in pyannote.audio. According to this page Open weights cc-by-4.0 pyannote-audio [View model](https://savrn.com/models/wespeaker-voxceleb-resnet34-lm) Model · Speech recognition ### [speaker-diarization-community-1](https://savrn.com/models/speaker-diarization-community-1) [Pyannote](https://savrn.com/model-publishers/pyannote) This pipeline ingests mono audio sampled at 16kHz and outputs speaker diarization. - stereo or multi-channel audio files are automatically downmixed to mono by averaging the channels. - audio files sampled at a different rate are resampled to 16kHz automatically upon loading. The main improvements brought by Community-1 are: - improved speaker assignment and counting - simpler reconciliation with transcription timestamps with exclusive speaker diarization - easy offline use (i.e. without internet connection) - (optionally) hosted on pyannoteAI cloud 1. pip install pyannote.audio 3. Create access token at hf.co/settings/tokens. Out of the box, Community-1 is much better than… Access requested at publisher cc-by-4.0 pyannote-audio [View model](https://savrn.com/models/speaker-diarization-community-1) Model · Speech recognition ### [parakeet-tdt-0.6b-v3](https://savrn.com/models/parakeet-tdt-0-6b-v3) [MLX Community](https://savrn.com/model-publishers/mlx-community) This model was converted to MLX format from nvidia/parakeet-tdt-0.6b-v3 using the conversion script. Please refer to original model card for more details on the model. Open weights cc-by-4.0 627M parameters mlx [View model](https://savrn.com/models/parakeet-tdt-0-6b-v3) Model · Text classification ### [twitter-roberta-base-sentiment-latest](https://savrn.com/models/twitter-roberta-base-sentiment-latest) [Cardiff NLP](https://savrn.com/model-publishers/cardiffnlp) This is a RoBERTa-base model trained on ~124M tweets from January 2018 to December 2021, and finetuned for sentiment analysis with the TweetEval benchmark. The original Twitter-based RoBERTa model can be found here and the original reference paper is TweetEval. This model is suitable for English. 0 -> Negative; 1 -> Neutral; 2 -> Positive This sentiment analysis model has been integrated into TweetNLP. You can access the demo here. Open weights cc-by-4.0 514 tokens transformers [View model](https://savrn.com/models/twitter-roberta-base-sentiment-latest) Model · Speech recognition ### [parakeet-tdt-0.6b-v2](https://savrn.com/models/parakeet-tdt-0-6b-v2) [MLX Community](https://savrn.com/model-publishers/mlx-community) This model was converted to MLX format from nvidia/parakeet-tdt-0.6b-v2 using the conversion script. Please refer to original model card for more details on the model. Open weights cc-by-4.0 618M parameters mlx [View model](https://savrn.com/models/parakeet-tdt-0-6b-v2) Model · Speech recognition ### [parakeet-ctc-1.1b](https://savrn.com/models/parakeet-ctc-1-1b) [NVIDIA](https://savrn.com/model-publishers/nvidia) parakeet-ctc-1.1b is an ASR model that transcribes speech in lower case English alphabet. This model is jointly developed by NVIDIA NeMo and Suno.ai teams. It is an XXL version of FastConformer CTC [1] (around 1.1B parameters) model. See the model architecture section and NeMo documentation for complete architecture details. To train, fine-tune or play with the model you will need to install NVIDIA NeMo. We recommend you install it after you've installed latest PyTorch version. There are several ways to use this model. Choose the one that fits your needs. NeMo-Speech.cpp provides a lightweight native C++ runtime for local inference with this model. After installing the runtime: See the… Open weights cc-by-4.0 1.1B parameters nemo [View model](https://savrn.com/models/parakeet-ctc-1-1b) Model · Audio classification ### [audiobox-aesthetics](https://savrn.com/models/audiobox-aesthetics) [AI at Meta](https://savrn.com/model-publishers/facebook) This model has been pushed to the Hub using the PytorchModelHubMixin integration: Unified automatic quality assessment for speech, music, and sound. Paper arXiv / MetaAI. Blogpost ai.meta.com This repository requires Python 3.9 and Pytorch 2.2 or greater. To install, you can clone this repo and run: if you only want to predict aesthetic scores from certain timestamp and save it as input.jsonl If you haven't downloade the checkpoint, the script will try to download it automatically. Otherwise, you can provide the path by --ckpt /path/to/checkpoint.pt If you have SLURM, run the following command Please adjust CPU & GPU settings using --slurm-gpu, --slurm-cpu depending on your nodes. 3. Output… Open weights cc-by-4.0 104M parameters [View model](https://savrn.com/models/audiobox-aesthetics) Model · Question answering ### [roberta-base-squad2](https://savrn.com/models/roberta-base-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is the roberta-base model, fine-tuned using the SQuAD2.0 dataset. It's been trained on question-answer pairs, including unanswerable questions, for the task of Extractive Question Answering. We have also released a distilled version of this model called deepset/tinyroberta-squad2. It has a comparable prediction quality and runs at twice the speed of deepset/roberta-base-squad2. Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over… Open weights cc-by-4.0 124M parameters 514 tokens transformers [View model](https://savrn.com/models/roberta-base-squad2) Model · Text generation ### [vlt5-base-keywords](https://savrn.com/models/vlt5-base-keywords) [VoiceLab.ai](https://savrn.com/model-publishers/voicelab) Results on demo model (different generation method, one model per language): Keywords generated with vlT5-base-keywords: encoder-decoder architecture, vlT5, keyword generation, scientific articles corpus The biggest advantage is the transferability of the vlT5 model, as it works well on all domains and types of text. The downside is that the text length and the number of keywords are similar to the training data: the text piece of an abstract length generates approximately 3 to 5 keywords. It works both extractive and abstractively. Longer pieces of text must be split into smaller chunks, and then propagated to the model. The model was trained on a POSMAC corpus. Polish Open Science… Open weights cc-by-4.0 275M parameters transformers [View model](https://savrn.com/models/vlt5-base-keywords) Model · Question answering ### [deberta-v3-base-squad2](https://savrn.com/models/deberta-v3-base-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is the deberta-v3-base model, fine-tuned using the SQuAD2.0 dataset. It's been trained on question-answer pairs, including unanswerable questions, for the task of Question Answering. Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. Timo Möller: timo.moeller [at] deepset.ai deepset is the company behind the production-ready open-source AI framework Haystack. We also… Open weights cc-by-4.0 184M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-base-squad2) Model · Question answering ### [bert-large-uncased-whole-word-masking-squad2](https://savrn.com/models/bert-large-uncased-whole-word-masking-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is a berta-large model, fine-tuned using the SQuAD2.0 dataset for the task of question answering. Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. deepset is the company behind the production-ready open-source AI framework Haystack. We also have a Discord community open to everyone! Open weights cc-by-4.0 335M parameters 512 tokens transformers [View model](https://savrn.com/models/bert-large-uncased-whole-word-masking-squad2) Model · Text to speech ### [tts-1.6b-en_fr](https://savrn.com/models/tts-1-6b-en-fr) [Kyutai](https://savrn.com/model-publishers/kyutai) See also the pre-print research paper, the project page, the Colab example, the GitHub repository, and the repository of voices. This is a model for streaming text-to-speech (TTS). Unlike offline text-to-speech, where the model needs the entire text to produce the audio, our model starts to output audio as soon as the first few words from the text have been given as input. This model is actually 1.8B parameters, not 1.6B as the name might suggest. The model architecture is a hierarchical Transformer that consumes tokenized text and generateds audio tokenized by Mimi, see the Moshi paper. The frame rate is 12.5 Hz and each audio frame is represented by 32 audio tokens, although you can use… Open weights cc-by-4.0 moshi [View model](https://savrn.com/models/tts-1-6b-en-fr) Model · Question answering ### [tinyroberta-squad2](https://savrn.com/models/tinyroberta-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is the distilled version of the deepset/roberta-base-squad2 model. This model has a comparable prediction quality and runs at twice the speed of the base model. This model was distilled using the TinyBERT approach described in this paper and implemented in haystack. Firstly, we have performed intermediate layer distillation with roberta-base as the teacher which resulted in deepset/tinyroberta-6l-768d. Secondly, we have performed task-specific distillation with deepset/roberta-base-squad2 as the teacher for further intermediate layer distillation on an augmented version of SQuADv2 and then with deepset/roberta-large-squad2 as the teacher for prediction layer distillation. Haystack is… Open weights cc-by-4.0 82M parameters 514 tokens transformers [View model](https://savrn.com/models/tinyroberta-squad2) Model · Translation ### [opus-mt-tc-big-en-tr](https://savrn.com/models/opus-mt-tc-big-en-tr) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) Neural machine translation model for translating from English (en) to Turkish (tr). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. You can also use OPUS-MT models with the transformers pipelines, for example: The work is supported by the European Language Grid as pilot… Open weights cc-by-4.0 1,024 tokens transformers [View model](https://savrn.com/models/opus-mt-tc-big-en-tr) Model · Translation ### [opus-mt-tc-big-tr-en](https://savrn.com/models/opus-mt-tc-big-tr-en) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) Neural machine translation model for translating from Turkish (tr) to English (en). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. You can also use OPUS-MT models with the transformers pipelines, for example: The work is supported by the European Language Grid as pilot… Open weights cc-by-4.0 235M parameters 1,024 tokens transformers [View model](https://savrn.com/models/opus-mt-tc-big-tr-en) Model · Translation ### [opus-mt-tc-big-ko-en](https://savrn.com/models/opus-mt-tc-big-ko-en) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) Neural machine translation model for translating from Korean (ko) to English (en). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. - More information about released models for this language pair: OPUS-MT kor-eng README - Tatoeba Translation… Open weights cc-by-4.0 209M parameters 1,024 tokens transformers [View model](https://savrn.com/models/opus-mt-tc-big-ko-en) Model · Question answering ### [bert-base-cased-squad2](https://savrn.com/models/bert-base-cased-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is a BERT base cased model trained on SQuAD v2 Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. deepset is the company behind the production-ready open-source AI framework Haystack. We also have a Discord community open to everyone! Open weights cc-by-4.0 108M parameters 512 tokens transformers [View model](https://savrn.com/models/bert-base-cased-squad2) Model · Tabular classification ### [TabSTAR](https://savrn.com/models/tabstar) [Alan Arazi](https://savrn.com/model-publishers/alana89) To fit a pretrained TabSTAR model to your own dataset, install the package: Paper: TabSTAR: A Foundation Tabular Model With Semantically Target-Aware Representations Open weights cc-by-4.0 47M parameters [View model](https://savrn.com/models/tabstar) Model · Question answering ### [xlm-roberta-base-squad2](https://savrn.com/models/xlm-roberta-base-squad2) [Deepset](https://savrn.com/model-publishers/deepset) Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. Evaluated on the SQuAD 2.0 dev set with the official eval script. "exact": 33.67279167589108 "total": 4517 "exact": 48.739495798319325 "total": 1190 Timo Möller: timo.moeller [at] deepset.ai deepset is the company behind the production-ready open-source AI framework Haystack. We also have a Discord community open to… Open weights cc-by-4.0 277M parameters 514 tokens transformers [View model](https://savrn.com/models/xlm-roberta-base-squad2) Model · Translation ### [opus-mt-tc-big-en-bg](https://savrn.com/models/opus-mt-tc-big-en-bg) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) Neural machine translation model for translating from English (en) to Bulgarian (bg). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. You can also use OPUS-MT models with the transformers pipelines, for example: The work is supported by the European Language Grid as pilot… Open weights cc-by-4.0 238M parameters 1,024 tokens transformers [View model](https://savrn.com/models/opus-mt-tc-big-en-bg) Model · Question answering ### [minilm-uncased-squad2](https://savrn.com/models/minilm-uncased-squad2) [Deepset](https://savrn.com/model-publishers/deepset) Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. Evaluated on the SQuAD 2.0 dev set with the official eval script. Timo Möller: timo.moeller@deepset.ai deepset is the company behind the production-ready open-source AI framework Haystack. We also have a Discord community open to everyone! Open weights cc-by-4.0 33M parameters 512 tokens transformers [View model](https://savrn.com/models/minilm-uncased-squad2) Model · Question answering ### [xlm-roberta-large-squad2](https://savrn.com/models/xlm-roberta-large-squad2) [Deepset](https://savrn.com/model-publishers/deepset) Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. Evaluated on the SQuAD 2.0 English dev set with the official eval script. Evaluated on German MLQA: test-context-de-question-de.json Evaluated on German XQuAD: xquad.de.json For doing QA at scale (i.e. many docs instead of single paragraph), you can load the model also in haystack: Timo Möller: timo.moeller@deepset.ai… Open weights cc-by-4.0 560M parameters 514 tokens transformers [View model](https://savrn.com/models/xlm-roberta-large-squad2) Model · Image and text to text ### [ProcVLM-2B](https://savrn.com/models/procvlm-2b) [Youhe Feng](https://savrn.com/model-publishers/ce-amtic) ProcVLM-2B is a procedure-grounded vision-language model for estimating progress rewards from robot manipulation observations. Given a task description and a recent window of video frames, the model reasons about the remaining atomic actions and predicts the current task completion percentage. ProcVLM-2B is designed for research on robot learning, progress reward modeling, embodied evaluation, and procedure-aware video understanding. Typical use cases include: - estimating task completion progress from robot videos; - producing dense progress rewards from sparse demonstrations; - adapting progress prediction to a new environment with one-shot LoRA fine-tuning. This model is not intended to… Open weights cc-by-4.0 2.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/procvlm-2b) Model · Question answering ### [roberta-large-squad2](https://savrn.com/models/roberta-large-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is the roberta-large model, fine-tuned using the SQuAD2.0 dataset. It's been trained on question-answer pairs, including unanswerable questions, for the task of Question Answering. Please note that we have also released a distilled version of this model called deepset/roberta-base-squad2-distilled. The distilled model has a comparable prediction quality and runs at twice the speed of the large model. Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering… Open weights cc-by-4.0 354M parameters 514 tokens transformers [View model](https://savrn.com/models/roberta-large-squad2) Model · Question answering ### [deberta-v3-large-squad2](https://savrn.com/models/deberta-v3-large-squad2) [Deepset](https://savrn.com/model-publishers/deepset) This is the deberta-v3-large model, fine-tuned using the SQuAD2.0 dataset. It's been trained on question-answer pairs, including unanswerable questions, for the task of Extractive Question Answering. Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. Evaluated on the SQuAD 2.0 dev set with the official eval script. deepset is the company behind the production-ready… Open weights cc-by-4.0 434M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-large-squad2) Model · Question answering ### [electra-base-squad2](https://savrn.com/models/electra-base-squad2) [Deepset](https://savrn.com/model-publishers/deepset) Evaluated on the SQuAD 2.0 dev set with the official eval script. Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. Vaishali Pal vaishali.pal [at] deepset.ai Timo Möller: timo.moeller [at] deepset.ai deepset is the company behind the production-ready open-source AI framework Haystack. We also have a Discord community open to everyone! Open weights cc-by-4.0 109M parameters 512 tokens transformers [View model](https://savrn.com/models/electra-base-squad2) Model · Question answering ### [bert-base-uncased-squad2](https://savrn.com/models/bert-base-uncased-squad2) [Deepset](https://savrn.com/model-publishers/deepset) Haystack is an AI orchestration framework to build customizable, production-ready LLM applications. You can use this model in Haystack to do extractive question answering on documents. To load and run the model with Haystack: For a complete example with an extractive question answering pipeline that scales over many documents, check out the corresponding Haystack tutorial. - Timo Möller: timo.moeller [at] deepset.ai deepset is the company behind the production-ready open-source AI framework Haystack. We also have a Discord community open to everyone! Open weights cc-by-4.0 109M parameters 512 tokens transformers [View model](https://savrn.com/models/bert-base-uncased-squad2) Model · Feature extraction ### [MicroGlot](https://savrn.com/models/microglot) [Athan Zhaohong Li](https://savrn.com/model-publishers/athanzli) A taxonomy-informed sparse DNA foundation model for microbial genomics. MicroGlot is a 23-layer decoder-only mixture-of-experts transformer pretrained on 378.3 billion nucleotides from 3.70 million sequences across 99,700 microbial species, spanning bacteria, archaea, fungi, protists, viruses and plasmids. It encodes the taxonomic hierarchy as hyperbolic (Poincaré) embeddings and uses them both as an input token and to steer expert routing. Paper: A Taxonomy-Informed Sparse DNA Foundation Model for Microbial Genomics\ A species is known if it is one of the 99,700 pretraining species (check with tokenizer.hasspecies(name)). For taxonomic classification tasks, use MicroGlot-plain, as… Open weights cc-by-4.0 3B parameters transformers [View model](https://savrn.com/models/microglot) Model · Audio classification ### [nocturne-v1-teacher](https://savrn.com/models/nocturne-v1-teacher) [Stratus Labs](https://savrn.com/model-publishers/stratus-labs) Nocturne is a bioacoustic species classifier trained by Stratus Labs covering the taxa that BirdNET and Perch don't: insects, amphibians, non-bird mammals, and reptiles. It's the "night side" of the soundscape — the taxa that are dominantly nocturnal or crepuscular, whose acoustic signal is the half of biodiversity monitoring bird-focused models leave behind. v1.1 (this revision) uses a 3× higher learning rate than v1 during backbone fine-tuning. Independent runs at both LRs plateaued around macro-F1 0.10; v1.1 edged past by ~3 %. Prior v1 weights are preserved in the commit history if you want to pin to revision=. Live public endpoint hosted on Stratus Labs infra (Apple M4 Max… Open weights cc-by-4.0 88M parameters transformers [View model](https://savrn.com/models/nocturne-v1-teacher) Model · Question answering ### [rai](https://savrn.com/models/rai) [Prayas Abhinav](https://savrn.com/model-publishers/prayasabhinav) What it is. An extractive question-answering model. Given a question and a short typed answer, it returns the phrase in the answer that answers the question, or nothing. It cannot generate text. It is the reading part of rai, a tool that says only whether a description is complete under a notion of completeness written down in advance. Every decision after the reading is made in plain code. Trained from. deepset/minilm-uncased-squad2 (CC-BY-4.0), deepset's fine-tune of Microsoft's MiniLM-L12-H384-uncased (MIT) on SQuAD 2.0. Same architecture, 33M parameters, nothing added. Credit to deepset and Microsoft. 0.3.7 trains on 0.3.6's rows and 5,800 more written in Claude Code. In 0.3.6's rows… Open weights cc-by-4.0 33M parameters 512 tokens transformers [View model](https://savrn.com/models/rai) Model · Audio classification ### [nocturne-v1.2d-teacher](https://savrn.com/models/nocturne-v1-2d-teacher) [Stratus Labs](https://savrn.com/model-publishers/stratus-labs) Audio Spectrogram Transformer, ~86M parameters, 2,196 output columns. This is the first candidate in the Nocturne line to clear our pre-committed release gate against the v1 teacher. It got there by fixing two supervision bugs, not by adding data — which is the most useful thing in this card. Read the limitations section before deploying it. On an unseen recording site this model is substantially worse than v1, and our much smaller distilled student beats it outright on the very test it was released for. Both are stated with numbers below. Identical 13,710-clip held-out test set, per-class thresholds calibrated on validation, scored over the 2,182 name-aligned core classes. Rows are… Open weights cc-by-4.0 88M parameters transformers [View model](https://savrn.com/models/nocturne-v1-2d-teacher) Model · Speech recognition ### [wav2vec2-xls-r-300m-cv7-turkish](https://savrn.com/models/wav2vec2-xls-r-300m-cv7-turkish) [Muhammet Poyraz](https://savrn.com/model-publishers/mpoyraz) This ASR model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on Turkish language. The following datasets were used for finetuning: - Common Voice 7.0 TR All validated split except test split was used for training. To support both of the datasets above, custom pre-processing and loading steps was performed and wav2vec2-turkish repo was used for that purpose. The following hypermaters were used for finetuning: - learningrate 2e-4 - numtrainepochs 10 - warmupsteps 500 - freezefeatureextractor - masktimeprob 0.1 - maskfeatureprob 0.05 - featprojdropout 0.05 - attentiondropout 0.05 - finaldropout 0.05 - activationdropout 0.05 - perdevicetrainbatchsize 8 - perdeviceevalbatchsize 8… Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/wav2vec2-xls-r-300m-cv7-turkish) Model · Translation ### [opus-mt-en-de](https://savrn.com/models/opus-mt-en-de) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) This model can be used for translation and text-to-text generation. CONTENT WARNING: Readers should be aware this section contains content that is disturbing, offensive, and can propagate historical and current stereotypes. Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Further details about the dataset for this model can be found in the OPUS readme: en-de Open weights cc-by-4.0 512 tokens transformers [View model](https://savrn.com/models/opus-mt-en-de) Model · Translation ### [opus-mt-ru-en](https://savrn.com/models/opus-mt-ru-en) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) This model can be used for translation and text-to-text generation. CONTENT WARNING: Readers should be aware this section contains content that is disturbing, offensive, and can propagate historical and current stereotypes. Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Further details about the dataset for this model can be found in the OPUS readme: ru-en Open weights cc-by-4.0 512 tokens transformers [View model](https://savrn.com/models/opus-mt-ru-en) Model · Image classification ### [AI-image-detector](https://savrn.com/models/ai-image-detector) [Matthew Maybe](https://savrn.com/model-publishers/umm-maybe) NOTE: Unless you are trying to detect imagery generated using older models such as VQGAN+CLIP, please use the updated version of this detector instead. This model is a proof-of-concept demonstration of using a ViT model to predict whether an artistic image was generated using AI. It was created in October 2022, and as such, the training data did not include any samples generated by Midjourney 5, SDXL, or DALLE-3. It still may be able to correctly identify samples from these more recent models due to being trained on outputs of their predecessors. Furthermore the intended scope of this tool is artistic images; that is to say, it is not a deepfake photo detector, and general computer imagery… Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/ai-image-detector) Model · Speech recognition ### [canary-180m-flash-gguf](https://savrn.com/models/canary-180m-flash-gguf) [Handy](https://savrn.com/model-publishers/handy-computer) GGUF conversions of nvidia/canary-180m-flash for use with transcribe.cpp. Ported from upstream commit pinned 2026-05-08. Validated against the NeMo reference at transcribe.cpp commit Offline multilingual speech-to-text and translation. A 182M-parameter multitask AED with a 17-layer FastConformer encoder and a 4-layer Transformer decoder. Supports automatic speech recognition in English, German, Spanish, and French, and bidirectional EN↔{DE, ES, FR} translation. Takes a 16 kHz mono WAV and produces a transcript. Not a streaming model; word/segment timestamps are upstream-experimental and not exposed in the v1 port. WER on the full LibriSpeech test-clean split (2,620 utterances), batch size… Open weights cc-by-4.0 transcribe.cpp [View model](https://savrn.com/models/canary-180m-flash-gguf) Model · Translation ### [opus-mt-zh-en](https://savrn.com/models/opus-mt-zh-en) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) This model can be used for translation and text-to-text generation. CONTENT WARNING: Readers should be aware this section contains content that is disturbing, offensive, and can propagate historical and current stereotypes. Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Further details about the dataset for this model can be found in the OPUS readme: zho-eng helsinkigitsha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 transformersgitsha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b portmachine: brutasse porttime: 2020-08-21-14:41 srcmultilingual: False tgtmultilingual: False reflen: 82826.0 brevitypenalty… Open weights cc-by-4.0 512 tokens transformers [View model](https://savrn.com/models/opus-mt-zh-en) Model · Speech recognition ### [canary-1b-v2-gguf](https://savrn.com/models/canary-1b-v2-gguf) [Handy](https://savrn.com/model-publishers/handy-computer) GGUF conversions of nvidia/canary-1b-v2 for use with transcribe.cpp. Ported from upstream commit pinned 2026-05-08. Validated against the NeMo reference at transcribe.cpp commit Offline multilingual speech-to-text and translation across 25 European languages. A 978M-parameter multitask AED with a 32-layer FastConformer encoder and an 8-layer Transformer decoder. Supports automatic speech recognition for any of the 25 supported languages, plus translation between supported language pairs (per the upstream model card). Takes a 16 kHz mono WAV and produces a transcript. Not a streaming model; word and segment timestamps from the upstream model are not exposed in the v1 port. WER on the full… Open weights cc-by-4.0 transcribe.cpp [View model](https://savrn.com/models/canary-1b-v2-gguf) Model · Translation ### [opus-mt-tc-big-ar-en](https://savrn.com/models/opus-mt-tc-big-ar-en) [Helsinki-NLP Research Group](https://savrn.com/model-publishers/helsinki-nlp) Neural machine translation model for translating from Arabic (ar) to English (en). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. You can also use OPUS-MT models with the transformers pipelines, for example: The work is supported by the European Language Grid as pilot… Open weights cc-by-4.0 1,024 tokens transformers [View model](https://savrn.com/models/opus-mt-tc-big-ar-en) Model · Speech recognition ### [canary-1b-flash-gguf](https://savrn.com/models/canary-1b-flash-gguf) [Handy](https://savrn.com/model-publishers/handy-computer) GGUF conversions of nvidia/canary-1b-flash for use with transcribe.cpp. Ported from upstream commit pinned 2026-05-08. Validated against the NeMo reference at transcribe.cpp commit Offline multilingual speech-to-text and translation. An 883M-parameter multitask AED with a 32-layer FastConformer encoder and a 4-layer Transformer decoder. Supports automatic speech recognition in English, German, Spanish, and French, and bidirectional EN↔{DE, ES, FR} translation. Takes a 16 kHz mono WAV and produces a transcript. Not a streaming model; word/segment timestamps are upstream-experimental and not exposed in the v1 port. WER on the full LibriSpeech test-clean split (2,620 utterances), batch size 1… Open weights cc-by-4.0 transcribe.cpp [View model](https://savrn.com/models/canary-1b-flash-gguf) Model · Speech recognition ### [owsm_ctc_v4_1B](https://savrn.com/models/owsm-ctc-v4-1b) [ESPnet](https://savrn.com/model-publishers/espnet) News: Our OWSM v4 paper won the Best Student Paper Award at INTERSPEECH 2025! Open Whisper-style Speech Model (OWSM) is the first fully open Whisper-style speech foundation model. It reproduces and advances OpenAI's Whisper-style training using publicly available data and open-source toolkits. The code, pre-trained model weights, and training logs are publicly released to promote open science in speech foundation models. OWSM-CTC (Peng et al., ACL 2024) is a novel encoder-only speech foundation model based on hierarchical multi-task self-conditioned CTC. It supports multilingual speech recognition, speech translation, and language identification within a single non-autoregressive model.… Open weights cc-by-4.0 espnet [View model](https://savrn.com/models/owsm-ctc-v4-1b) Model · Speech recognition ### [parakeet-tdt-0.6b-v3](https://savrn.com/models/parakeet-tdt-0-6b-v3-3) [LiteRT Community (FKA TFLite)](https://savrn.com/model-publishers/litert-community) Fully on-device, 25-language speech-to-text in a single file. This is (600M-parameter FastConformer encoder + Token-and-Duration Transducer, TDT, decoder) packaged as a.litertlm bundle for LiteRT-LM, Google's open-source on-device inference runtime, alongside standalone.tflite and AOT-compiled NPU variants for LiteRT. Everything needed to transcribe arbitrary-length audio with automatic language detection, punctuation, and capitalization is inside the one.litertlm file: INT8 encoder, hybrid stateful/stateless TDT decoder, unified 8,192-token multilingual SentencePiece tokenizer, and Speech2TextMetadata (16 kHz resampler and 128-bin Slaney log-mel front-end configuration) — no Python and no… Open weights cc-by-4.0 litert-lm [View model](https://savrn.com/models/parakeet-tdt-0-6b-v3-3) Model · Zero shot image classification ### [DermLIP_ViT-B-16](https://savrn.com/models/dermlip-vit-b-16) [Siyuan Yan](https://savrn.com/model-publishers/redlessone) DermLIP is a vision-language model for dermatology, trained on the Derm1M dataset—the largest dermatological image-text corpus to date. python import openclip from PIL import Image import torch model,, preprocess = openclip.createmodelandtransforms( 'hf-hub:redlessone/DermLIPViT-B-16' model.eval() tokenizer = openclip.gettokenizer('hf-hub:redlessone/DermLIPViT-B-16') image = preprocess(Image.open("yourskinimage.png")).unsqueeze(0) PADCLASSNAMES = [ "nevus", "basal cell carcinoma", "actinic keratosis", "seborrheic keratosis", "squamous cell carcinoma", "melanoma" template = lambda c: f'This is a skin image of {c}' text = tokenizer([template(c) for c in PADCLASSNAMES]) with torch.nograd()… Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/dermlip-vit-b-16) Model · Translation ### [plt5-base](https://savrn.com/models/plt5-base) [Allegro ML Research](https://savrn.com/model-publishers/allegro) plT5 models are T5-based language models trained on Polish corpora. The models were optimized for the original T5 denoising target. plT5 was trained on six different corpora available for Polish language: The training dataset was tokenized into subwords using a sentencepiece unigram model with vocabulary size of 50k tokens. If you use this model, please cite the following paper: The model was trained by Machine Learning Research Team at Allegro and Linguistic Engineering Group at Institute of Computer Science, Polish Academy of Sciences. You can contact us at: klejbenchmark@allegro.pl Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/plt5-base) Model · Translation ### [plt5-large](https://savrn.com/models/plt5-large) [Allegro ML Research](https://savrn.com/model-publishers/allegro) plT5 models are T5-based language models trained on Polish corpora. The models were optimized for the original T5 denoising target. plT5 was trained on six different corpora available for Polish language: The training dataset was tokenized into subwords using a sentencepiece unigram model with vocabulary size of 50k tokens. If you use this model, please cite the following paper: The model was trained by Machine Learning Research Team at Allegro and Linguistic Engineering Group at Institute of Computer Science, Polish Academy of Sciences. You can contact us at: klejbenchmark@allegro.pl Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/plt5-large) Model · Text to speech ### [pocket-tts-coreml](https://savrn.com/models/pocket-tts-coreml) [Fluid Inference Community](https://savrn.com/model-publishers/fluidinference) CoreML conversion of kyutai/pocket-tts for on-device inference on Apple platforms. 4 pre-encoded voices in constantsbin/: - alba (default), azelma, cosette, javert Voice cloning is supported: the Mimi encoders are included (mimiencoder.mlmodelc at the root for English, v2.1/ /mimiencoderv3.mlmodelc per language pack). In FluidAudio, use PocketTtsManager.cloneVoice() or fluidaudiocli tts "…" --backend pocket --clone-voice speaker.wav. Core ML packs for Kyutai's six small English models released 2026-10-20: english18m2026-10, english7m2026-10, english2m2026-10 and their reward-fine-tuned refl2026-10 twins (a different architecture from the 100M: grouped-query attention, zproj, low-rank text… Open weights cc-by-4.0 coreml [View model](https://savrn.com/models/pocket-tts-coreml) Model · Text generation ### [semsimula-ladder-live-owt-d384-l2-attention-potential-rfqk](https://savrn.com/models/semsimula-ladder-live-owt-d384-l2-attention-potential-rfqk) [Dimitar Gueorguiev](https://savrn.com/model-publishers/dimitarpg13) Settled is the mean of the last three evals (52.83, 53.05, 52.81); best 50.97 at step 31,000, final 52.81. Consecutive evals near the end differ by about ±2 PPL. - The hardening recovers what the unhardened field lost: 2.4% better than the same model without it. - The hardened field still adds nothing measurable: 0.4% better than the model without the field, which is within eval noise. The three L=2 models with the register path settle within 2.5% of one another, whatever else is switched on. - Without the registers the same field is worth much more: the conservative-only model with this field and no register read settles at 54.27, 6.0% below the conservative-only model. The field and the… Open weights cc-by-4.0 pytorch [View model](https://savrn.com/models/semsimula-ladder-live-owt-d384-l2-attention-potential-rfqk) Model · Reinforcement learning ### [VFIG-4B-GGUF](https://savrn.com/models/vfig-4b-gguf) [Team Mradermacher](https://savrn.com/model-publishers/mradermacher) static quants of https://huggingface.co/XunmeiLiu/VFIG-4B For a convenient overview and download list, visit our model page for this model. weighted/imatrix quants are available at https://huggingface.co/mradermacher/VFIG-4B-i1-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… Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/vfig-4b-gguf) Model · Tabular regression ### [nordic-price-forecast](https://savrn.com/models/nordic-price-forecast) [Elpriser.org](https://savrn.com/model-publishers/elpriser) Day-ahead electricity price forecasts 2 to 9 days out, hourly, in EUR/MWh, for every bidding zone in Denmark, Norway, Sweden, Finland and the Netherlands: Retrained and re-uploaded daily, so the weights here are what production runs. A companion model specialised to Denmark alone lives at A single LightGBM quantile model covers all 13 zones, with zone as a categorical feature. The zones share nearly all their structure — the same calendar effects, the same weather physics, the same lag behaviour — so pooling lets a thin zone borrow strength from a busy one, and leaves one thing to maintain instead of thirteen. Pooling only works because the model can identify a zone's price level before it… Open weights cc-by-4.0 lightgbm [View model](https://savrn.com/models/nordic-price-forecast) Model ### [marian-mt-bbc-en-pcm](https://savrn.com/models/marian-mt-bbc-en-pcm) [National Information Technology Development Agency Hub AI Research Team](https://savrn.com/model-publishers/nithub-ai) Parallel Sentences from the message translation (English) and Pidgin translation of the Bible. - Pidgin sentences from BBC Pidgin and English translations by GPT3.5-turbo TBA Bridging the Gap with Retrieval-Augmented Generation: Making Prosthetic Device User Manuals Available in Marginalised Languages Open weights cc-by-4.0 512 tokens transformers [View model](https://savrn.com/models/marian-mt-bbc-en-pcm) Model · Speech recognition ### [parakeet-tdt-0.6b-v3](https://savrn.com/models/parakeet-tdt-0-6b-v3-2) [Cactus Compute, Inc.](https://savrn.com/model-publishers/cactus-compute) Parakeet TDT 0.6b is a speech recognition model optimized for on-device performance and live transcription. Open weights cc-by-4.0 [View model](https://savrn.com/models/parakeet-tdt-0-6b-v3-2) Model ### [parakeet](https://savrn.com/models/parakeet) [Volantis Labs](https://savrn.com/model-publishers/volantis-labs) Speech recognition: Parakeet TDT 0.6B v3 with ternary encoder weights, as a Core ML model set and an ONNX model set. - coreml/: Core ML compiled models (Preprocessor, Encoder with 2-bit weights, Decoder, JointDecision; the joint also returns its float32 logits), with config.json, tokens.txt and SHA256SUMS. macOS 15 or later. - onnx/: ONNX encoder (2-bit MatMulNBits), decoder, joiner and voice activity model, with config.json, tokens.txt and SHA256SUMS. Core ML model set on the GPU, on a 62.9-minute subset of the AMI meeting corpus: 22.49% word error rate, about 126x real time on an Apple M3 Max. Open weights cc-by-4.0 [View model](https://savrn.com/models/parakeet) Model · Audio classification ### [pyannote-speaker-diarization.axera](https://savrn.com/models/pyannote-speaker-diarization-axera) [ZY](https://savrn.com/model-publishers/hy-2012) pyannote/speaker-diarization-community-1（cc-by-4.0）说话人日志模型的 AX650N 量化部署包： 输入 16 kHz 单声道会议音频，输出 RTTM 说话人时序标签。 - 模型: 8 个 axmodel（U16 激活 / S8 权重） - 依赖: Python 仅 numpy + scipy + scikit-learn + soundfile + axengine；C++ 为预编译 aarch64 可执行文件 常用参数：--step 2.0（滑动步长，2.5 更快 +0~1.5 pp）、--num-speakers 0（0=自动估计）。 切换 NPU2 版：--model-dir modelsnpu2（runax650.sh 通过环境变量 MODELDIR 指定）。 - 输入: 16 kHz 单声道 wav；输出: RTTM（SPEAKER 1 SPEAKERXX ） - 流水线: sincnet 分割（im2col + MatMul）→ 4 层 BiLSTM（96 步 cell，FP32 状态跨块） → 说话人嵌入（ResNet34 + TSTP）→ PLDA/AHC/VBx 聚类 - InstanceNorm / TSTP 统计池化在 U16 量化域数值崩坏，放主机 FP32 计算 - 完整转换源码见 GitHub: pyannote-speaker-diarization.AXERA 指标表所用 AMI / AliMeeting / VoxConverse 三个测试集的下载链接、参考处理与评分脚本见 ¹… Open weights cc-by-4.0 [View model](https://savrn.com/models/pyannote-speaker-diarization-axera) Model · Text generation ### [alea-r4-3B-A0.7B-Base-ckpts](https://savrn.com/models/alea-r4-3b-a0-7b-base-ckpts) [ALEA Institute](https://savrn.com/model-publishers/alea-institute) Qwen3-MoE architecture; 3.32B parameters in total (3,315,557,632, tied input/output embeddings), 673M active per token. These are intermediate base-model pretraining checkpoints, not instruction-tuned. Safety and behaviour status: pending final checks. They are published for research: to follow what the model learns over training and to load any checkpoint with stock transformers. They are not a release. R4 is ALEA's next model after R3, trained only on lawful data: U.S. legal, regulatory and financial public-record text from the KL3M Data Project, plus synthetic, dictionary, knowledge and task data created by ALEA. main holds only this card. Each checkpoint is its own branch; load one with… Open weights cc-by-4.0 transformers [View model](https://savrn.com/models/alea-r4-3b-a0-7b-base-ckpts) Model · Speech recognition ### [Parakeet-TDT-0.6B-V3-coreml](https://savrn.com/models/parakeet-tdt-0-6b-v3-coreml) [Glimpse Dictation](https://savrn.com/model-publishers/glimpse-dictation) This repository contains the optional Apple Silicon package for Parakeet TDT 0.6B V3 (GGUF) in Glimpse. It requires the transcribe.cpp Parakeet Core ML adapter with decoder-only GGUF support. The application dependency integration is under development; these are not standalone transcription executables. The combined download is 1.11 GB, without duplicating the encoder weights in the GGUF. For CPU/GPU transcription on Windows or Mac, use instead. The Core ML package requires Apple Silicon. The model originates from NVIDIA Parakeet TDT 0.6B V3, licensed CC BY 4.0. Both artifacts derive from Handy's parakeet-tdt-0.6b-v3-Q80.gguf using transcribe.cpp's scripts/convert-parakeet-gguf-to-coreml.py… Open weights cc-by-4.0 coreml [View model](https://savrn.com/models/parakeet-tdt-0-6b-v3-coreml) Model · Speech recognition ### [parakeet-tdt-0.6b-v2](https://savrn.com/models/parakeet-tdt-0-6b-v2-2) [Cactus Compute, Inc.](https://savrn.com/model-publishers/cactus-compute) Parakeet TDT 0.6b v2 is an English-only speech recognition model optimized for on-device performance and live transcription. Open weights cc-by-4.0 [View model](https://savrn.com/models/parakeet-tdt-0-6b-v2-2) Model · Speech recognition ### [parakeet-tdt-ctc-110m-glade](https://savrn.com/models/parakeet-tdt-ctc-110m-glade) [Josh Leverette](https://savrn.com/model-publishers/coder543) Glade is a native Swift library for on-device speech inference on Apple silicon. This repository supplies converted model assets and configuration for Glade's model-specific host runtime. Glade's runtime source is currently private. Apple Neural Engine conversions of NVIDIA's Parakeet TDT–CTC 110M. Each Core AI profile supports TDT and CTC decoding from the same encoder, with native token and word timings. A portable Core ML graph bundle is included as a bonus, qualified on Mac, iPhone and Apple Watch. The English model has a 17-layer, 512-wide FastConformer encoder, a one-layer 640-wide LSTM TDT predictor, and a separate CTC head. These conversions preserve the original weights without… Open weights cc-by-4.0 [View model](https://savrn.com/models/parakeet-tdt-ctc-110m-glade) Model · Speech recognition ### [parakeet-ultra-spoken-coreml](https://savrn.com/models/parakeet-ultra-spoken-coreml) [MimicScribe](https://savrn.com/model-publishers/mimicscribe) A Core ML build of Parakeet Ultra whose decoder and joint were fine-tuned to write numbers, symbols and abbreviations the way they are spoken ("twenty five percent", "point nine", "December thirty first") instead of mixing digits and words. The encoder, preprocessor, tokenizer and file layout are unchanged from drop-in replacement: only Decoder.mlmodelc/weights/weight.bin and the joint weights (JointDecision.mlmodelc / JointDecisionv3.mlmodelc, byte-identical) differ. Built for MimicScribe, which transcribes overlapping audio windows and stitches them; one consistent written form for numbers makes the overlaps agree. - moondream/parakeet-ultra — post-training of the above, CC BY 4.0… Open weights cc-by-4.0 coreml [View model](https://savrn.com/models/parakeet-ultra-spoken-coreml) Model · Text to speech ### [vctk_tts_train_speechbrain_xvector_vits](https://savrn.com/models/vctk-tts-train-speechbrain-xvector-vits) [ESPnet](https://savrn.com/model-publishers/espnet) This model was trained by ftshijt using vctk recipe in espnet. Follow the ESPnet installation instructions if you haven't done that already. Open weights cc-by-4.0 espnet [View model](https://savrn.com/models/vctk-tts-train-speechbrain-xvector-vits) Model · Text to speech ### [vctk_tts_train_espnet_rawnet_vits](https://savrn.com/models/vctk-tts-train-espnet-rawnet-vits) [ESPnet](https://savrn.com/model-publishers/espnet) This model was trained by ftshijt using vctk recipe in espnet. Follow the ESPnet installation instructions if you haven't done that already. Open weights cc-by-4.0 espnet [View model](https://savrn.com/models/vctk-tts-train-espnet-rawnet-vits) ## Questions ### Can I use Creative Commons Attribution 4.0 models commercially? Yes. CC BY 4.0 permits sharing and adapting the work, including commercially, provided the creator is credited and changes are indicated. ### Which Creative Commons Attribution 4.0 models are most downloaded? By monthly downloads reported by the Hugging Face Hub: wespeaker-voxceleb-resnet34-LM (5.8M); speaker-diarization-community-1 (5M); parakeet-tdt-0.6b-v3 (2.5M). ## Other Licenses - [apache-2.0](https://savrn.com/models/licenses/apache-2-0) 3395 - [mit](https://savrn.com/models/licenses/mit) 1335 - [cc-by-nc-4.0](https://savrn.com/models/licenses/cc-by-nc-4-0) 215 - [bsd-3-clause](https://savrn.com/models/licenses/bsd-3-clause) 108 - [gemma](https://savrn.com/models/licenses/gemma) 63 - [cc-by-nc-sa-4.0](https://savrn.com/models/licenses/cc-by-nc-sa-4-0) 56 - [llama3.2](https://savrn.com/models/licenses/llama3-2) 40 - [unknown](https://savrn.com/models/licenses/unknown) 33 - [agpl-3.0](https://savrn.com/models/licenses/agpl-3-0) 32 - [llama3.1](https://savrn.com/models/licenses/llama3-1) 29 - [cc-by-sa-4.0](https://savrn.com/models/licenses/cc-by-sa-4-0) 20 - [gpl-3.0](https://savrn.com/models/licenses/gpl-3-0) 20 [See all](https://savrn.com/models/licenses)