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.
Updated 2026-10-19 · How the library is built
11,442 models, sorted by most downloaded.
This is an ONNX version of nlp-chula/phayathaibert-thai-pos-tagger, a PhayaThaiBERT model fine-tuned for Universal POS tagging on UD Thai-TUD. It runs with onnxruntime and tokenizers, without PyTorch or Transformers. This repository only converts the model. All credit for training goes to the original authors at nlp-chula and to the PhayaThaiBERT authors. Made for the PyThaiNLP POS tagging engine Input is a list of pre-segmented Thai words. Each word is tagged with the prediction for its first subword, the same alignment the model was trained with. Keep each model call at 415 subwords or fewer, counting and. Split longer input into sentences, or into chunks of whole words, and tag each one…
This is an experimental Mae codebase for Classification. It keeps the tiny setup intentionally manageable so architecture changes can be inspected before a full training run. - The Python file contains the model and runnable example or training entry point. - config.json records the generated architecture settings. - trainingargs.json records the default experiment recipe. - model.safetensors is a valid initialization checkpoint for smoke tests; it is not presented as a trained benchmark checkpoint. - No benchmark score is claimed in this repository. The included configuration uses sgd with a polynomial schedule. These are starting values in the script, not evidence of a completed run. For…
A specialized 3.8B biomedical & clinical reasoning foundation model built on Microsoft's Phi-4-mini-instruct, formatted for standard Hugging Face transformers and PyTorch. 1. Stage 1 (STEM Foundation): 116,000 instruction pairs across NCERT Classes 6–12 (Physics, Chemistry, Biology) eliminating foundational science hallucinations. 2. Stage 2 (PubMed 2026 Evidence): 12 recent 2026 clinical update archives from NCBI FTP covering survival outcomes (OS, PFS, HR), targeted therapeutics, and clinical trial endpoints. 3. Stage 3 (Comprehensive Internal Medicine): Balanced multi-specialty clinical curriculum (cardiology, nephrology, endocrinology, pulmonology) with an active oncology replay buffer.…
A specialized 3.8B biomedical & clinical reasoning model built on Microsoft's Phi-4-mini-instruct, optimized natively for Apple Silicon Metal acceleration via Apple MLX. The model underwent a 3-stage transfer learning curriculum: 1. Stage 1 (STEM Foundation): 116,000 instruction pairs across NCERT Classes 6–12 (Physics, Chemistry, Biology) eliminating foundational science hallucinations. 2. Stage 2 (PubMed 2026 Evidence): 12 recent 2026 clinical update archives from NCBI FTP covering survival outcomes (OS, PFS, HR), targeted therapeutics, and clinical trial endpoints. 3. Stage 3 (Comprehensive Internal Medicine): Balanced multi-specialty clinical curriculum (cardiology, nephrology…
This is an ONNX version of Horizon-Labs/phishing-scam-detector-base. It was automatically converted and uploaded using this Hugging Face Space. See the pipeline documentation for text-classification: https://huggingface.co/docs/transformers.js/api/pipelines#modulepipelines.TextClassificationPipeline Flags phishing, scams and spam in emails, SMS, chat and social-media messages in many languages. Three labels: - fraud: phishing (fake bank, delivery, account or tax messages with links), scams (prizes, investment and crypto schemes, fake jobs, "hi mum, new number", romance, advance fee), requests for passwords, one-time codes or payment details, malware lures; - spam: unwanted advertising and…
A fine-tuned DistilBERT sequence classifier for binary email triage: legitimate and spam. It is a prototype trained and evaluated on a historical SpamAssassin corpus. It is not a phishing or malware detector and should not be treated as a complete email security system. Use as an experimental text-only aid to sort historical or otherwise appropriately reviewed email into ham/spam queues. Route uncertain, empty, or unsupported inputs to human review. Do not use the output as the sole basis for blocking senders, taking account actions, or making other consequential decisions. This model does not authenticate senders, inspect attachments, resolve or visit links, verify SPF/DKIM/DMARC, or…
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. Three experimental conversions of Fermion Research's Phonon-2 for Apple silicon and Core AI APIs on macOS/iOS 27. LUT6 is recommended for speed. The other variants expose the tradeoff between download size, working memory, and inference time. Phonon-2 is an encoder-heavy FastConformer/TDT model derived from NVIDIA Parakeet V3. The conversions preserve its trained five-value encoder weights and original INT6 decoder values. They use FP16 activations and the…
This model is a custom PyTorch U-Net architecture (SketchUNet) trained to convert RGB photos into target outline sketches using the rhfeiyang/photo-sketch-pair-500 dataset. - Combined Sketch Loss (L1 + Sobel Edge Loss): 0.14555 - Mean Absolute Error (L1 / MAE): 0.04962 This model converts an RGB image into a grayscale sketch using a PyTorch U-Net architecture. The model checkpoint is available as: Once the model architecture is defined, inference can be reduced to
Dose-conditioned networks trained on the photon-thinning bridge, whose forward process removes detected photons by binomial thinning, so every state of the process is distributed as a scan acquired at a lower dose. One checkpoint covers a calibrated range of doses: the network takes a low-dose reconstruction together with the dose coordinate t and predicts the full-dose image. bridge2detect.pt and bridgeldct.pt produce the simulated-dose results of the paper; bridgeldct.pt is also the model evaluated on the official Mayo reduced-dose projections. bridge2detectfano.pt is trained at the Fano-corrected effective counts Ieff = I0 / F and is the model evaluated on the 3.3%-current 2DeteCT scans.…
A π0 policy (openpi, JAX) fine-tuned with LoRA (gemma2blora + gemma300mlora) from the openpi pi0base checkpoint, on the RoboSynChallenge/cobotmagicSimitemshandover dataset (LeRobot v2.1). 3e28642c63555fcfc64412b2cd83d6358bb54330a3c5713b5d6774b49d6fca43 (18 files); normalization statistics cae91c7b9df3ff7a87f8caa67967e7ec609e70b8952dcff44a561098036c5135; training save record RESEARCHSAVE label=23999 step=24000 optstate=1966a406d1976f52. 2.5e-6; AdamW, batch 32, seed 42; action offset +1; prompt = the dataset's task text), then continued from step 16000 to 24000 with --resume on the same data, the same schedule (learning rate about 1.314e-5 at step 16000 and about 4.79e-6 at step 24000), the…
Evaluation-ready Orbax checkpoint from pi0 full fine-tuning on washdish50.
Native OpenPI JAX/Orbax checkpoint (params/, assets/), config pi05axisjoint, camera0 RGB (+ wrist camera where the evaluator provides it) + 9D joint state → 9D absolute joint targets. Weights are subject to the Gemma Terms of Use; OpenPI license applies to the code.
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.