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 2026-10-22 · How the library is built
12,283 models, sorted by most downloaded.
RSR-27B is Qwen-3.8-27B finetuned with Recursive Self-Rewrite (RSR) trajectories.
Rainbow-AI-Specialist is a custom multi-domain fine-tuned AI model based on google/gemma-2-2b-it, trained using QLoRA via Hugging Face TRL (SFTTrainer). You can load and test this adapter easily in Python using PEFT and Transformers: import torch from transformers import AutoModelForCausalLM, AutoTokenizer from peft import PeftModel modelid = "google/gemma-2-2b-it" tokenizer = AutoTokenizer.frompretrained(modelid) basemodel = AutoModelForCausalLM.frompretrained(modelid, torchdtype=torch.float16, devicemap="auto") model = PeftModel.frompretrained(basemodel, "RainbowLord/Rainbow-AI-Specialist") inputs = tokenizer("### User: Write a Python script for a simple trading bot.\n### Rainbow:"…
RavenX-CyberAgent-Qwen3.6-35B-A3B converted to the Splash schema-4 splash-packed-q4-moe format for Apple Silicon. This repository is a Splash runtime package, not an MLX checkpoint, GGUF file, or Transformers checkpoint. The target and tokenizer come from the which derives from the The compatible DFlash 2 draft and vision weights are reused from All 40 converted language-model layer files were built from RavenX's weights. The reused official Splash weight assets are the draft model and vision encoder. Install Splash on an Apple Silicon Mac, then replace YOURUSERNAME with this repository's owner: After Splash prints Ready, open http://127.0.0.1:8000 or send a chat completion request. The 8K…
Single-stream, 50 Hz speech coding at 650 / 800 bps. This model repository contains the complete ReLMCodec inference release: both codec checkpoints, their matching YAML configs, the frozen W2v-BERT 2.0 parameters embedded in each checkpoint, and the code needed to reconstruct audio. The weights have no optimizer, scheduler, discriminator, or RNG state. The 8K checkpoint is the 200K perceptual-stage export. The 64K checkpoint is the available 50K additional fine-tune, selected by Log-Mel/PESQ/STOI on a separate 50-file dev set. The closest retained record to the paper's 64K table used an unavailable 30K fine-tune weight. Checkpoint provenance records this distinction and the actual…
ReScraper is a 0.6B refiner that replaces the heuristic HTML-to-text stack of a pretraining data pipeline — a rule-based scraper followed by rule-based cleaning filters — with a single small language model. The model reads the visible text of a raw web page, rendered one block per line with line ids, and generates a short program: 1. An block of line removals (rm 1-4, rm 6, …) that strips headers, menus, boilerplate and other non-content lines. 2. One operation tag for the page: An executor applies the program to the page, so extraction and cleaning happen in one model instead of a cascade of separate stages. The system prompt and the executor that applies the generated program are in the…
把 Real-ESRGAN 的 RealESRGANx4plus （RRDBNet，23 個 RRDB block，16.7 M 參數）轉成 OpenVINO IR，並用 NNCF 做成 INT4 / INT8 / FP16 多種精度版本。所有模型都是純 OpenVINO 推論， 執行時不需要 PyTorch，只要 openvino + opencv + numpy（首次使用會自動從本 repo 下載權重）。 測試機：Intel Xeon Platinum 8559C（Emerald Rapids，16 vCPU，含 AMX-INT8 / AVX512-VNNI / AVX512-FP16）、 OpenVINO 2026.4.1、tile=128、tilepad=16。 畫質（3 張 inputs/ 測試圖，PSNR / SSIM 對 fp16 版本；fp16 與 PyTorch FP32 差異 直接下載 00sourcelr.mp4 你的瀏覽器不支援 video 標籤，請改用 直接下載 20sidebyside1920x780.mp4 若播放器不支援 HTML video，可直接看這支 GIF： 原始 PyTorch FP32（放大後） 你的瀏覽器不支援 video 標籤，請改用 直接下載 10pytorchfp324x.mp4 OpenVINO FP16 你的瀏覽器不支援 video 標籤，請改用 直接下載 11openvinofp164x.mp4 OpenVINO INT8（W8A8） 你的瀏覽器不支援 video 標籤，請改用 直接下載…
This repository hosts two checkpoints of the Realtime-Venus system: - Realtime-Venus-Omni (Realtime-Venus-Omni/): the 9B audio-visual interaction model. It continuously watches and listens, decides whether and when to respond, and generates text and speech on a shared causal timeline. Adapted from MiniCPM-o 4.5, it supports proactive interaction, semantic interruption handling, and training-free long-video memory. - Realtime-Venus-Audio (Realtime-Venus-Audio/): the audio-focused checkpoint on the same streaming backbone, for audio understanding and audio-driven conversation with text or speech output. Both directories contain model weights and custom Hugging Face Transformers code. The…
The Realtime-Venus-Omni checkpoint from inclusionAI/Realtime-Venus (itself a fine-tune of openbmb/MiniCPM-o-45), with refusal behaviour reduced by abliteration of the language tower. BF16, same file layout and remote code as upstream. An NVFP4 build is at Only the language tower (llm., Qwen3-8B shape, 36 layers). The vision tower, audio tower, resampler, audio projection, TTS talker and token2wav assets are upstream's, unmodified. - 72 matrices were changed: selfattn.oproj and mlp.downproj in all 36 layers. - The largest relative change is ‖ΔW‖/‖W‖ = 0.379. The upstream sample media (assets/sample1real.mp4, assets/speechin.mp4) are not included. - The language tower was extracted as a…
quantised with NVIDIA ModelOpt 0.46.1. It is 9.1 GB on disk. The abliteration details and numbers are on that page. In short: heretic ARA branch, ARA-LoRA rank 128, trial 83, refusals 98 → 4 of 100, KL 0.249 on recheck, no healing applied. - llm calibration: 256 samples × 512 tokens of tatsu-lab/alpaca (shuffled, seed 0), run through the extracted text tower. Mean NLL on 8 calibration chats: BF16 3.941, NVFP4 fake-quant 4.065 (+0.123). - vpm calibration: 80 images from johnowhitaker/imagenette2-320. Three quarters were unsliced; one quarter used maxslicenums=4. - vpm held-out check (16 images): - One RTX 5060 Ti 16 GB (SM120) and one RTX 3050 Ti Laptop 4 GB (SM86), with lmhead on CPU during…
Diffusers pipelines for Rectified-CFG++, a training-free guidance method for rectified-flow models (NeurIPS 2025). This repo has no weights. It holds pipeline code that you load on top of an existing checkpoint with custompipeline. Classifier-free guidance (CFG) extrapolates from the unconditional velocity toward the conditional one at every ODE step. On rectified-flow models this extrapolation pushes samples off the learned data manifold, and at high guidance scales the images show over-saturated colours and broken structure. Rectified-CFG++ replaces that extrapolation with a predictor-corrector update: 1. Predictor: take a step along the conditional velocity from the current point. 2.…
OpenArm 양팔 로봇(biopenarmfollower, 카메라 3대, state·action 16차원)의 Redox 시연 200개로 체크포인트 하나가 "task 문장 → 서브태스크 문장 생성 → 그 문장을 조건으로 액션 청크(50스텝)"를 한다. task 문장 (학습에 쓴 것은 이것 하나다): 1. lift the metal plate out of the left beaker 2. move the plate over the blue solution 3. point at the plate with the left hand 4. lower the plate into the blue solution 5. hold the plate still in the solution 6. raise the plate and point at it 7. put the plate into the left beaker 8. return to the home posture - lerobot PR #4184의 브랜치 codex/pi052-language-policy가 필요하다 (학습에 쓴 커밋 3298e15, lerobot 0.6.2). PyPI의 0.6.1에는 pi052 정책도 --interactive도 없다. - 그 커밋의 액션 추론은 그대로는 죽는다: TypeError: PI05Pytorch.embedprefix() takes 5 positional…
This is a trained model of a Reinforce agent playing Pixelcopter-PLE-v0. To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction
This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.
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.