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 2026-10-18 · How the library is built
11,122 models, sorted by most downloaded.
This repository contains a fine-tuned XLM-RoBERTa Base model for 3-class sentiment classification in Sindhi. The model predicts one of the following labels: - positive - neutral - negative This model is intended for research, experimentation, and lightweight production prototyping on Sindhi-language sentiment analysis tasks. It may be useful for short-text classification such as reviews, feedback, and social-media snippets. It should not be used as the only basis for high-stakes decisions. Performance can vary by domain, dialect, and writing style, and human review is recommended when errors could affect people or organizations. The saved model artifacts in this repository report the…
SNIF (StateGate Neural Inference Fabric) is an open-source composite causal language model that dynamically connects Qwen2.5-0.5B (Edge Prefill and Fast Decode) and DeepSeek-R1-1.5B (Cognitive Reasoning) via a continuous 2048-dimensional latent interconnect. Note: Research paper and extended theoretical evaluation in preparation.
Snowflake/snowflake-arctic-embed-l-v2.0 converted to GGUF with llama.cpp b10630. Embeddings have 1024 dimensions. With Shabti, in custommodels.ini
在 Dexmal/DM05 上做 LoRA 微调得到的 SO-101 单臂策略： 仿真与真机混成一份，共 3698 集 / 1,280,507 帧 / 10 个任务，全部 30 fps、640×480： - LoRA（all-linear），可训参数 324M / 6.15B = 5.27% - 动作模式 RELATIVE，动作块长度 50 - 等效 batch 48（4 卡 × 每卡 6 × 梯度累积 2），5000 步，bf16，cosine 学习率 成败取环境自带的 issuccess（物体在料箱内、夹爪松开、物体与机械臂都静止），每局上限 500 步： 需要 dexbotic 框架。适配器里记录的底座是 Dexmal/DM05， 完整的数据准备、训练与仿真评测流程见 dexbotic 的 docs/so101.md，教学示例见 适配器随底座 Dexmal/DM05 采用 Gemma 许可。
底座里的 VAE、文本编码器与 Wan2.2-TI2V-5B 扩散主干仍从 Dexmal/DW05-Robotwin 取，本仓只放微调后的权重。 - 起点：Dexmal/DW05-Robotwin，剔掉与双臂机器人绑定的动作头与本体编码器（宽度对不上 SO-101，重新学） - 全量微调 + DeepSpeed ZeRO-1，每卡 batch 1 × 3 卡，20000 步，学习率 1e-4 cosine - SO-101 的 6 维状态填进 RobotWin 的 16 槽排布：前五个槽放五个臂关节，夹爪落在第 7 槽；三路视图用 顶视 + 腕部 + 腕部 - 本体状态在训练时没有归一化：推理时把 6 维状态排进 16 槽、补一位终止标志 0 后原样送入，不要套 RobotWin 的分位数归一化。 用法与评测脚本见 dexbotic 的 docs/so101.md（script/so101/dw05simcheck.py）， Apache-2.0，随底座 Dexmal/DW05-Robotwin。
Inference checkpoint from an early Stack3 fine-tune of a two-camera SO-101 π0.5 policy. It is warm-started from the Stack2 checkpoint that performed well in the user's real-robot trials, followed by 4,000 optimizer updates on the collected Stack3 full-chain dataset. There is no held-out validation set in this initial run. Training health or loss does not establish successful three-block stacking. This checkpoint has not been verified on the physical robot; it is provided for deployment testing, not as a claim of task success. The repository contains policy weights, model configuration, tokenizer, and matching observation/action pre- and post-processors with the normalization statistics…
GR00T N1.7 is an open, cross-embodiment foundation model from NVIDIA for generalized humanoid robot reasoning and skills. It uses a Cosmos-Reason2/Qwen3-VL backbone and a flow-matching action transformer to predict actions conditioned on vision, language, and proprioception. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot groot guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and calibrate your robot and…
Action Chunking with Transformers (ACT) is an imitation-learning method that predicts short action chunks instead of single steps. It learns from teleoperated data and often achieves high success rates. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot act guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and calibrate your robot and cameras. - Record data & train a policy — the end-to-end imitation-learning…
SmolVLA is a compact, efficient vision-language-action model that achieves competitive performance at reduced computational costs and can be deployed on consumer-grade hardware. This policy has been trained and pushed to the Hub using LeRobot. Learn how to train and run it in the LeRobot smolvla guide, or browse the full documentation. The policy consumes these observation features and produces these action features. Inputs Outputs New to LeRobot? These guides cover the full workflow: - Install LeRobot — set up the lerobot package. - Hardware setup — assemble, wire, and calibrate your robot and cameras. - Record data & train a policy — the end-to-end imitation-learning walkthrough. - CLI…
Independent further-training of the new sim-only 20,000-step policy on 10 real10 episodes for 5,000 additional steps. Both real10 and Cosmos10 branches initialize from the same sim-only parent; neither continues from the other branch. - Native parent processors and statistics retained, with stock clipping; no unclipping ablation. - Visual tuning enabled, language tuning disabled, BF16 autocast with FP32 parameters; image augmentation enabled. - Checkpoints at 2,500 and 5,000 additional steps. This repository contains the final 5,000-step continuation. The native policy and processor files must remain together. See training-provenance.json for parent weight hash and exact provenance.…
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.