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.
11,442Models
3,247Datasets
285Papers
6,976Publishers
12,091Sourced relationships
Updated 2026-10-19 · How the library is built
11,442 models, sorted by most downloaded.
This repository contains a 500-cluster K-means codebook for Whisper large-v3-turbo encoder representations, trained on Japanese audio from Lambda Audio Corpus. Each 1,280-dimensional audio frame can be assigned to its nearest cluster center to produce a discrete unit ID from 0 to 499. These IDs can be used as targets for speech representation learning. They are learned acoustic units, not text tokens or labeled phonemes. The codebook was trained on 18,000,004 sampled frames from 199.89 hours of accepted audio. The repository includes the cluster centers, training features, and metadata. Original audio and transcripts are not included. The files are stored under codebook/. Only centroids.npy…
whisper-heb-v7-multi (Hebrew / English / Arabic / Russian / Yiddish with built-in language ID) plus speech-free negatives: the model has learned to answer with nothing on line noise, hold music, tones and silence instead of inventing words. Same architecture (Whisper large-v3-turbo, 809M), same languages, same usage; drop-in replacement for v7. On telephony audio Whisper fine-tunes hallucinate short phrases when there is no speech. v7 produced a transcript for 98% of real line-noise clips (27% of them "תודה רבה") and 46% of synthetic noise. v8 was warm-started from v7 and trained 5,000 steps on the v7 corpus plus 83,920 speech-free rows (real Voicenter line noise ×4, synthetic…
Hindi audio in → Roman-script Hinglish captions out. 100% in-browser, no server. This is a browser-fixed repack of (fine-tune of openai/whisper-tiny for Hindi → Hinglish). Original model and training credit belong to the original author — this repo only fixes the two browser blockers and trims the package for web use. Young India reads Hindi in Roman script (WhatsApp/Instagram culture) — they speak Hindi but don't read Devanagari. This model outputs captions the ~102 MB total — small enough to run on-device via transformers.js, including mobile browsers. 1. tokenizer.json — BPE merges were stored as lists (["a", "b"]); converted to strings ("a b"), which is what the browser tokenizer path…
Model · Speech recognition
whisper-large-v3-turbo-anv-sot-ct2
A CTranslate2 conversion of a Whisper large-v3-turbo fine-tune for Sesotho (Southern Sotho) trained by DSFSI on the African Next Voices corpus. The weights are unchanged apart from the format conversion and int8 quantization, for use with All credit for the model goes to the original authors. Please cite and refer to the source model and the African Next Voices dataset. Whisper has no Sesotho language token. The fine-tune decodes under the English token (en), which Whisper's language detection narrowly picks for it, so pass it explicitly to avoid detection drifting to another language. Turning off conditiononprevioustext avoids repetition loops on long audio. Converted from source revision…
Model · Speech recognition
whisper-large-v3-turbo-anv-sot-mlx
An MLX conversion of a Whisper large-v3-turbo fine-tune for Sesotho (Southern Sotho) trained by DSFSI on the African Next Voices corpus. The weights are unchanged apart from the format conversion, for use with Apple silicon. All credit for the model goes to the original authors. Please cite and refer to the source model and the African Next Voices dataset. Whisper has no Sesotho language token. The fine-tune decodes under the English token (en), which Whisper's language detection narrowly picks for it, so pass it explicitly to avoid detection drifting to another language. Turning off conditiononprevioustext avoids repetition loops on long audio. Converted from source revision…
Model · Speech recognition
whisper-large-v3-turbo-anv-zul-ct2
A CTranslate2 conversion of a Whisper large-v3-turbo fine-tune for Zulu trained by DSFSI on the African Next Voices corpus. The weights are unchanged apart from the format conversion and int8 quantization, for use with All credit for the model goes to the original authors. Please cite and refer to the source model and the African Next Voices dataset. Whisper has no isiZulu language token. The fine-tune decodes under the English token (en), which Whisper's language detection narrowly picks for it, so pass it explicitly to avoid detection drifting to another language. Turning off conditiononprevioustext avoids repetition loops on long audio. Converted from source revision…
Model · Speech recognition
whisper-large-v3-turbo-anv-zul-mlx
An MLX conversion of a Whisper large-v3-turbo fine-tune for Zulu trained by DSFSI on the African Next Voices corpus. The weights are unchanged apart from the format conversion, for use with Apple silicon. All credit for the model goes to the original authors. Please cite and refer to the source model and the African Next Voices dataset. Whisper has no isiZulu language token. The fine-tune decodes under the English token (en), which Whisper's language detection narrowly picks for it, so pass it explicitly to avoid detection drifting to another language. Turning off conditiononprevioustext avoids repetition loops on long audio. Converted from source revision…
Model · Speech recognition
whisper-medium_indic_voices_63097_ratio_30_70_lr_1e-4_lgid_None_hinglish_mixed_scripts_FT
This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 16 - evalbatchsize: 16 - gradientaccumulationsteps: 2 - totaltrainbatchsize: 32 - lrschedulertype: constantwithwarmup - lrschedulerwarmupsteps: 500 - numepochs: 100 - Transformers 5.0.0.dev0 - Pytorch 2.9.0+cu126 - Datasets 3.6.0 - Tokenizers 0.22.2
Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation. Trained on 680k hours of labelled data, Whisper models demonstrate a strong ability to generalise to many datasets and domains without the need for fine-tuning. Whisper was proposed in the paper Robust Speech Recognition via Large-Scale Weak Supervision by Alec Radford et al from OpenAI. The original code repository can be found here. Disclaimer: Content for this model card has partly been written by the Hugging Face team, and parts of it were copied and pasted from the original model card. Whisper is a Transformer based encoder-decoder model, also referred to as a sequence-to-sequence model. It…
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 model is a fine-tuned version of openai/whisper-small on the generator dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 1e-05 - trainbatchsize: 8 - evalbatchsize: 8 - gradientaccumulationsteps: 2 - totaltrainbatchsize: 16 - lrschedulertype: linear - lrschedulerwarmupsteps: 300 - numepochs: 3.0 - Transformers 4.57.6 - Pytorch 2.2.1+cu121 - Datasets 5.0.1 - Tokenizers 0.22.2
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).
Unmodified copy of selected files from ggerganov/whisper.cpp at revision 5359861c739e955e79d9a303bcbc70fb988958b1, kept so that SubVox Pro (Unity editor plugin by ExNihilus) can download a stable, pinned version of the model. All credit goes to the original authors: the conversion by ggerganov and the original model openai/whisper-large-v3. - ggml-large-v2.bin - ggml-large-v3.bin - ggml-large-v3-turbo.bin
This whisper model was trained 2x faster with Unsloth and Huggingface's TRL library.
Core ML / WhisperKit pack converted from then 8-bit linear weight-quantized (AudioEncoder + TextDecoder) for smaller on-device footprint on Apple Silicon. (not "यार मैं कल ऑफिस नहीं जाऊँगा") Use language: "en" in WhisperKit decoding options. The fine-tune maps Hindi speech onto English/Latin tokens; forcing Hindi often breaks romanization. Typical layout (folder name may match what you uploaded): Or after Hub download, point --model / repo flags at this repository per current whisperkit-cli help. - 8-bit linear weight quantization is lossy storage; compute remains floating point. - Expect near-parity with the float Core ML build, not bit-identical transcripts. - Always spot-check on your…
colab1/latest.pt is overwritten as the run progresses; colab1/step.pt are milestones every 1000 steps. Load with the trainer (RESUMECKPT=), not as a standalone model.
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.