This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: - evalloss: 0.5162 - evalaccuracy: 1.0 - evalruntime: 20.8464 - evalsamplespersecond: 0.911 - evalstepspersecond: 0.144 The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - numepochs: 10 - mixedprecisiontraining: Native AMP - Transformers 5.18.0 - Pytorch 2.11.0+cpu - Datasets 4.8.5 - Tokenizers 0.23.2
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 model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - numepochs: 10 - mixedprecisiontraining: Native AMP - Transformers 5.18.0 - Pytorch 2.11.0+cu130 - Datasets 4.8.5 - Tokenizers 0.23.2
This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - numepochs: 10 - mixedprecisiontraining: Native AMP - Transformers 5.18.0 - Pytorch 2.11.0+cu130 - Datasets 4.8.5 - Tokenizers 0.23.2
This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 4 - evalbatchsize: 4 - lrschedulertype: linear - numepochs: 10 - mixedprecisiontraining: Native AMP - Transformers 5.18.0 - Pytorch 2.11.0+cpu - Datasets 4.8.5 - Tokenizers 0.23.2
This model is a fine-tuned version of Kkonjeong/wav2vec2-base-korean on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 0.0001 - trainbatchsize: 4 - evalbatchsize: 4 - lrschedulertype: linear - numepochs: 10 - mixedprecisiontraining: Native AMP - Transformers 5.18.0 - Pytorch 2.11.0+cpu - Datasets 4.8.5 - Tokenizers 0.23.2
\[Notice\] Preprint will be released soon! K-Fold predicts biomolecular complex structures and binding-induced conformational changes. Through an apo-to-holo diffusion bridge, K-Fold aims to capture conformational changes in systems such as G protein-coupled receptors (GPCRs). K-Fold supports proteins, DNA, RNA, small molecules, and chemical modifications without multiple sequence alignments (MSAs). This model repository provides K-Fold pretrained weights and configuration. Inference and training code, examples, and data preprocessing workflows are available on GitHub. K-Fold uses pretrained AtlasLM for protein sequence representations and TriProRep for protein structure representations.…
This repository provides the chemical component dictionary and pretrained encoder weights used by K-Fold. These assets are downloaded automatically when needed. - assets/ccd.pkl: Precomputed chemical component dictionary (CCD). - weights/protstructbbtokenizer.pth: Protein backbone tokenizer weights copied from TriProRep-3B. - weights/protstructfatokenizer.pth: Protein full-atom tokenizer weights copied from TriProRep-3B. - weights/protstructencoder3b.pth: Protein structure encoder weights converted to bfloat16 from TriProRep-3B-finetune. - weights/rnaseq1b.pth: Pretrained RNA sequence encoder weights. Source code for the protein structure encoder and tokenizers is available in the TriProRep…
A from-scratch hybrid OCR pipeline for the saeidseyfi/khattat dataset, built by saeidseyfi from 100% free, offline components — no training required, no paid APIs. 1. OpenCV preprocessing — upscale, Otsu/adaptive binarization, deskew, morphology 2. EasyOCR (fa+ar+en) — CRAFT detection + CRNN recognition 3. Tesseract 5 (tessdatabest: fas / ara / eng) — LSTM models with word confidences 4. Confidence ensemble — script detection + weighted candidate selection + language normalization 5. pix2tex (LaTeX-OCR) for math formulas — ViT -> LaTeX + math-token candidate scoring 6. OpenCV contour classification for hand-drawn shapes (10 classes) 7. Caption OCR — first ink cluster below the shape -> en…
Handwritten line recognizer for Persian / Arabic / English text plus math symbols, fine-tuned on the saeidseyfi/khattat dataset (line crops, input 32x256, max 88 label chars). CPU (2-core) budget fine-tune; decoder+head trained, STR backbone frozen The training pipeline is time-budgeted and fully resumable — continuing on a GPU with the same script converges substantially further. - best.pt -- model checkpoint (state dict + vocab) - vocab.txt -- 176-char vocabulary (fa / ar / en / digits / math) - demokhattatparseq.mp4 -- demo: test lines, ground truth vs prediction - evalreport.json, log.txt, worstsamples.json -- metrics
Three recovered line OCR checkpoints: CNN + BiLSTM + CTC (step 58,000), CNN + BiGRU + CTC (step 8,500), and CNN + Transformer Encoder + CTC (step 87,000). All weights load strictly and CPU inference was executed. Checkpoints contain weights and metadata; optimizer/RNG states were removed. They are for inference and weight initialization, not exact optimizer-state resumption. Open CompareModels.ipynb in Google Colab and Run all. No GPU or Hugging Face account is required. It displays images, expected answers and actual predictions. Optional own-image upload is disabled by default. The 20 demo images were selected from 40 screened printed validation examples because the Transformer predicted…
This model detects text lines and mathematical formula rows in scanned Cambodian textbooks for OCR line cropping and document layout analysis.
KHTST é um modelo multimodal compacto para PT-BR com roteamento por S-SOM, janela PARALELA de contexto (NMAXTOKENS 393.216), classificadores Jev auto-calibrados, o tronco causal híbrido MAMBA-3 5:1 (v12), otimização MuonClip + AdamW, a quantização GSQ (Gumbel-Softmax) com decaimento analítico de τ e restrição RCO no Stiefel com gradiente riemanniano no backward (v13) e — novidade v14 — a EXTENSÃO DO GSQ+RCO A TODOS OS com DECORRELAÇÃO DAS ÁRVORES provada, RCO nas NoPE com cota estrutural do logit (ℓmax ≤ d/√dh — QK-Clip estruturalmente INATIVO), GateGSQ nos mecanismos de atenção, ajuste analítico K/τmin (trade-off compressão× erro do codebook), percepção conjunta imagem ⊕ profundidade ⊕…
Kiel-3-Omni is an advanced, staged multimodal conversational model developed by KielTech. Built directly upon Kiel-2.1-Omni, Kiel-3-Omni is engineered with a modular training pipeline that natively integrates voice capabilities, robust text instruction-following, and upcoming vision understanding. This repository hosts the Phase 2 Unified Preview, combining core voice foundations with high-capacity text Supervised Fine-Tuning (SFT). You can load and run inference with the unified weights using the following snippet: Preview Notice This repository represents an intermediate preview build (v0.2-preview) under active development by KielTech. Performance and capabilities will expand further…
Learning Kinematic Frequency-Aware Disentanglement for Motion Style Transfer and Editing, SIGGRAPH Asia 2026. These are the five checkpoints used by the paper figures and the live demo. The content motion is encoded by the motion VAE. The style motion is split by the IMF extractor into high-, mid-, and low-frequency bands. High and mid bands supply the fine style. A summary of the low band is encoded by MotionCLIP into the coarse-style token. The dual-style denoiser writes a latent, and the VAE decoder turns it back into a motion. Use the five files together. About 2.3 GB in total. SHA256SUMS lists the SHA-256 and byte size of each file. CLIP ViT-B/32 is not in this repository. The clip…
Experimental NAS/pruning/full LoRA recovery artifact. Created using Bonsai by Prism ML. Derived from original SFT, revision 011ee6c8d05d49144fa5747bedf23e8904883564. - q1/base-Q10.gguf + q1/adapter-f16.gguf: 169895008 bytes combined. Cap applies to this deployment pair, not this whole repository. - model/: matching pruned HF base and tokenizer, necessary for this adapter. Do not attach the adapter to the unpruned original base. - hfadapter/: final R08 LoRA weights; metadata/: pruning map, recipe, certificates and provenance. - contract/: original trained contract including recommendations. benchmark/: OOD comparison, recommendations ignored only in core-v1 scoring. One full epoch, 77,358…
Experimental NAS/pruning/full LoRA recovery artifact. Created using Bonsai by Prism ML. Derived from original SFT, revision 011ee6c8d05d49144fa5747bedf23e8904883564. - q1/base-Q10.gguf + q1/adapter-f16.gguf: 197170592 bytes combined. Cap applies to this deployment pair, not this whole repository. - model/: matching pruned HF base and tokenizer, necessary for this adapter. Do not attach the adapter to the unpruned original base. - hfadapter/: final R08 LoRA weights; metadata/: pruning map, recipe, certificates and provenance. - contract/: original trained contract including recommendations. benchmark/: OOD comparison, recommendations ignored only in core-v1 scoring. One full epoch, 77,358…
A QLoRA adapter for OpenAI's gpt-oss-20b that turns it into KinShield's phone-call evidence detector. KinShield is an entry in the AWS "Zero to Shipped" hackathon. The model reads a call transcript and returns strict JSON with: - a risk level (LOW / MEDIUM / HIGH) and a score - every warning sign, with the exact quoted line, a timestamp, one of 7 fixed signals and a weight It answers directly on the harmony final channel, without writing out reasoning first. Status: this is a research result. The live KinShield app still uses the stock gpt-oss-20b on Amazon Bedrock. This adapter is not hosted anywhere public. The adapter matches gpt-oss-120b, a model about 6× larger, and edges out the stock…
A 22M-parameter phone-call scam classifier (MiniLM-L6, int8 ONNX, 22.9 MB) built for KinShield, an entry in the AWS "Zero to Shipped" hackathon. It reads a call transcript and returns a score from 0 to 1 for each of KinShield's 7 warning signs, plus a HIGH-risk probability. It runs on CPU in about 40–80 ms. What's new in v3: v3 was distilled from Solomonwilsonr/kinshield-gpt-oss-20b-lora, our fine-tuned gpt-oss-20b. v2 was distilled from the stock gpt-oss-20b. Both runs use int8 ONNX. v3 still misses scam016 (a Medicare card scam) and scam019 (a Hinglish UPI "beta in trouble" call), which it rates LOW, and rates scam014 MEDIUM. impersonation, emergency, secrecy, financialrequest…
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.