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.
12,579Models
3,583Datasets
291Papers
7,578Publishers
12,850Sourced relationships
Updated 2026-10-23 · How the library is built
12,579 models, sorted by most downloaded.
This is an MLX model for Apple Silicon. It is not directly compatible with the PyTorch CUDA runner. Converted from mlx-community/MOSS-TTS-8B-8bit at revision b9262754ad5245942472c7b4ff2b53cec3a07ec0: Q6 backbone and Q8 speech path, affine quantization with group size 64. Structurally validated on Linux; Apple Silicon runtime and speech-quality validation are pending. See RESULT.md, recipe.json, and the audit reports for details. The upstream model card follows. This MLX implementation brings MOSS-TTS to Apple Silicon through the MLX runtime, adapting the TTS backbone, multi-head audio token generation, and MOSS Audio Tokenizer workflow to run inside the MLX-Audio stack. As part of…
GGUF quantization of OpenMOSS-Team/MOSS-VL-Realtime-NF4 (11B cross-attention VLM: Qwen3 text trunk with 12 pure cross-attn layers + SigLIP tower with deepstack), repackaged for the SkyVision.cpp inference engine. - NF4 weights are dequantized bit-exactly against bitsandbytes CUDA first (checkmossvlnf4.py); - the fused cross-KV projection weight is built at load time (engine-side); 12 GB GPU deployments: --n-ctx 4096 (and MOSSXCTX=6144 for long videos). Full chat, multi-turn sessions, realtime streaming (proactive narration with ) and a web UI are available via --serve. Follows the base model's license (see the OpenMOSS-Team/MOSS-VL-Realtime-NF4 page).
A small causal Qwen3 model that predicts 32 MOSS v2 audio-codebook indices from the 12 indices of the current frame and up to ten previously generated HQ frames. It is trained from scratch on German, English, Spanish and French speech. This is an experimental enhancement model. Validation cross entropy measures token prediction under teacher forcing. It does not establish an improvement in perceived audio quality. Use the generated-history listening evaluation to judge speech content, speaker identity, artifacts and long-utterance stability. Root inference weights: step 64,699, selected by held-out extra-20-codebook cross entropy 5.818123. - Step 12,940: 100 originals / 200 input cases…
The data/ directory contains policy weights for MyoSuite/MSK-Bench tasks, together with their configuration files, training logs, and evaluation results. The top-level folders identify algorithms or experiment series. Weights from different folders are not interchangeable, even when their task names look similar. The word best is part of the original filenames; this README does not infer which metric was used to select those checkpoints. To reproduce an evaluation, also check the configuration, environment version, and any associated preprocessing or normalization state. This directory has three task folders. Each numbered subdirectory contains one experiment run. The environment names…
Inference checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos Download the base model separately: Qwen/Qwen3-VL-4B-Instruct. This repository contains inference modules, not a standalone base model. This variant uses the last 8 seconds as the current observation, represented by 8 frames. Follow docs/videoevaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime. Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark…
Inference checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos Download the base model separately: Qwen/Qwen3.5-0.8B. This repository contains inference modules, not a standalone base model. Follow docs/videoevaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime. Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark assets retain their respective upstream terms.
Inference checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos Download the base model separately: Qwen/Qwen3.5-2B. This repository contains inference modules, not a standalone base model. Follow docs/videoevaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime. Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark assets retain their respective upstream terms.
Inference checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Evaluation requires the MaLoW code release. The code has been prepared locally; public GitHub publication is pending. Intended repository: MaLoW (forthcoming). Download the base model separately: Qwen/Qwen3.5-2B. This repository contains inference modules, not a standalone base model. Follow docs/vlaevaluation.md in the MaLoW code release for Simpler WidowX. This checkpoint additionally requires the Qwen3.5-2B-MLP baseline adapter as the frozen policy base. Both bundles are required. Reported reference success rates (not a fresh evaluation of this export): datasetstatistics.json supplies action…
Inference checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos Download the base model separately: Qwen/Qwen3.5-4B. This repository contains inference modules, not a standalone base model. Follow docs/videoevaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime. Overall computed from listed subtask scores using benchmark weights 2500:1500:250:250. Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark assets retain…
Inference checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos Download the base model separately: Qwen/Qwen3.5-4B. This repository contains inference modules, not a standalone base model. Evaluation uses the last 4 seconds as the current observation, represented by 4 frames. Follow docs/videoevaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime. Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark assets…
Checkpoint for Memory as Weights: Internalizing Long-Term History for Streaming Videos. Paper: Memory as Weights: Internalizing Long-Term History for Streaming Videos Download the base model separately: Qwen/Qwen3.5-9B. This repository contains inference modules, not a standalone base model. Follow docs/videoevaluation.md in the MaLoW code release for OVO-Bench and StreamingBench. The JSON metadata records the inference configuration; load it through the MaLoW runtime. Weights and inference metadata are distributed under Apache-2.0. See LICENSE and NOTICE; base models and benchmark assets retain their respective upstream terms.
Try MadStickArt online → — generate a storyboard in your browser before downloading. Adjust composition and camera controls, then export clean or annotated PNGs and scene JSON. The playground runs the released neural decoder locally and uses optional rule-based scene suggestions; Strands Decider is available in the downloadable Python pipeline. MadStickArt is a small, trained 12,192-parameter image model for clean stick-figure storyboard scenes at 956 × 400 pixels, exactly 2.39:1. This first public release is danger-room/MadStickArt-v0.1.0, licensed under Apache 2.0. Pin revision v0.1.0 for reproducible downloads. The model is a custom PyTorch layout-conditioned neural raster decoder. A…
MadStickArt is a trained, 61,296-parameter neural raster decoder for stick-figure storyboard scenes. It reconstructs 956 × 400 grayscale frames from four explicit layout masks: actors, setting, props, and motion. Its output contract is 2.39:1. Try MadStickArt online → — the browser trial is deployed separately. Its deployed model/version and renderer are recorded in the Space's model-provenance.json. The renderer included with this release supports 186 settings and 320 props; the deployed trial has its own model and renderer pins and is updated through a separate reviewed stage. Browser suggestions use transparent rules, while Strands decisions run in the downloadable Python pipeline. The…
ALL PREVIEW IMAGES HAVE COMFYUI METADATA. MageTrail is a Danbooru/E621 proof of concept Full-Finetune of Microsoft's MageFlow 4B T2I model, now 2.8B custom architecture, using a diversity maximized condensed 41k images dataset as a way to tune booru concept and tags based prompting + Illustration capabilities into the model without having to tune with the full booru dataset. My second attempt at fine-tuning an image model on larger scale, this finetune aim to prove to the open source community on MageFlow 4B having good potential as an architecture for further finetuning. ~ V0.1 to V0.3 have spent 593 dollars and has ultimately shown the model potential in quickly learning and adapting…
Um modelo de linguagem de 1,4 milhão de parâmetros, treinado do zero em português do Brasil pela Mangaba AI. Relatório técnico completo (dados, arquitetura, Muon, ablações, controles e destilação): https://huggingface.co/spaces/mangaba-ai/mangaba-nano-artigo Não há modelo base: os pesos começaram aleatórios e o tokenizer também foi treinado do zero sobre o mesmo corpus. O treino rodou nas GPUs NVIDIA L40S da própria Mangaba, em datacenter no Brasil. É um modelo de pesquisa e de borda: cabe em 5,6 MB, roda em qualquer CPU e serve como base para estudar escalas minúsculas, fazer ajuste fino em tarefas estreitas (classificação, autocompletar, extração simples) ou rodar em microcontroladores e…
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.