全站搜索

AI 模型 449

• legal-bert-small-uncased↗

  nlpaueb

  LEGAL-BERT is a family of BERT models for the legal domain, intended to assist legal NLP research, computational law, and legal technology applications. To pre-train the different variations of LEGAL-BERT, we collected 12 GB of diverse English legal text from several fields (e.g., legislation, court cases, contracts) scraped from publicly available resources. Sub-domain variants (CONTRACTS-, EURLEX-, ECHR-) and/or general LEGAL-BERT perform better than using BERT out of the box for domain-specific tasks. This is the light-weight version of BERT-BASE (33% the size of BERT-BASE) pre-trained from scratch on legal data, which achieves comparable performance to larger models, while being much more efficient (approximately 4 times faster) with a smaller environmental footprint. I. Chalkidis, M. Fergadiotis, P. Malakasiotis, N. Aletras and I. Androutsopoulos. "LEGAL-BERT: The Muppets straight out of Law School". In Findings of Empirical Methods in Natural Language Processing (EMNLP 2020) (Short Papers), to be held online, 2020. (https://aclanthology.org/2020.findings-emnlp.261) \* LEGAL-BERT-BASE is the model referred to as LEGAL-BERT-SC in Chalkidis et al. (2020); a model trained from scratch in the legal corpora mentioned below using a newly created vocabulary by a sentence-piece tokenizer trained on the very same corpora.

• Wan2.1-I2V-14B-720P-gguf↗

  city96

  This is a direct GGUF conversion of Wan-AI/Wan2.1-I2V-14B-720P All quants are created from the FP32 base file, though I only uploaded FP16 due to it exceeding the 50GB max file limit and gguf-split loading not currently being supported in ComfyUI-GGUF. The model files can be used with the ComfyUI-GGUF custom node.

• ideogram_4_turbotime_lora↗

  ostris

  This LoRA will turn Ideogram 4 into a few step model. It is capable of generating good quality images in as little as 2 steps with no CFG and no unconditional model. It was created with a continuious turbo training method I am working on and still perfecting. I will release more information on that once it is fully polished. This will likely recieve updated versions as the method improves.

• Qwen3.5-35B-A3B-abliterated-GGUF↗

  HeYujie

  This repository contains the GGUF quantized versions of huihui-ai/Huihui-Qwen3.5-35B-A3B-abliterated huihui-ai/Huihui-Qwen3.5-35B-A3B-abliterated is an uncensored version of Qwen/Qwen3.5-35B-A3B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

• Qwen3-VL-8B-Heretic-1.3.0↗

  DreamFast

  An abliterated version of Qwen's Qwen3-VL-8B-Instruct created using Heretic v1.3.0. This model has reduced refusals while maintaining model quality. You can see the docker, scripts and configurations used to make these files on Heretic Docker Github. ComfyUI converted text encoders are provided in various formats, along with the original bf16 safetensors. This is a vision-language model (supports both text and image inputs). It does **not** support thinking/reasoning tokens.

• LFM2.5-2.6B-Base↗

  LiquidAI

  LFM2.5 is a new family of hybrid models designed for **on-device deployment**. It builds on the LFM2 architecture with extended pre-training and reinforcement learning. LFM2.5-2.6B-Base is the pre-trained text-only checkpoint, used to create all the LFM2.5-2.6B variants. It has the following features: This pre-trained checkpoint is only recommended for tasks that require heavy fine-tuning, like language-specific (e.g., Japanese) or domain-specific (e.g., medical) assistants, training on proprietary data, or experimenting with novel post-training approaches.

• coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k↗

  timm

  A timm specific CoAtNet (w/ a MLP Log-CPB (continuous log-coordinate relative position bias motivated by Swin-V2) image classification model. Pretrained in `timm` on ImageNet-12k (a 11821 class subset of full ImageNet-22k) and fine-tuned on ImageNet-1k by Ross Wightman. MaxxViT covers a number of related model architectures that share a common structure including: - CoAtNet - Combining MBConv (depthwise-separable) convolutional blocks in early stages with self-attention transformer blocks in later stages. - MaxViT - Uniform blocks across all stages, each containing a MBConv (depthwise-separable) convolution block followed by two self-attention blocks with different partitioning schemes (window followed by grid). - CoAtNeXt - A timm specific arch that uses ConvNeXt blocks in place of MBConv blocks in CoAtNet. All normalization layers are LayerNorm (no BatchNorm). - MaxxViT - A timm specific arch that uses ConvNeXt blocks in place of MBConv blocks in MaxViT. All normalization layers are LayerNorm (no BatchNorm). - MaxxViT-V2 - A MaxxViT variation that removes the window block attention leaving only ConvNeXt blocks and grid attention w/ more width to compensate. Aside from the major variants listed above, there are more subtle changes from model to model. Any model name with the string `rw` are `timm` specific configs w/ modelling adjustments made to favour PyTorch eager use. These were created while training initial reproductions of the models so there are variations. All models with the string `tf` are models exactly matching Tensorflow based models by the original paper authors with weights ported to PyTorch. This covers a number of MaxViT models. The official CoAtNet models were never released.

• gemma-3-270m↗

  unsloth

  Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. Gemma 3 models are multimodal, handling text and image input and generating text output, with open weights for both pre-trained variants and instruction-tuned variants. Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions. Gemma 3 models are well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as laptops, desktops or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone. These models were trained on a dataset of text data that includes a wide variety of sources. The 27B model was trained with 14 trillion tokens, the 12B model was trained with 12 trillion tokens, 4B model was trained with 4 trillion tokens, the 1B with 2 trillion tokens, and the 270M with 6 trillion tokens. The knowledge cutoff date for the training data was August 2024. Here are the key components: The combination of these diverse data sources is crucial for training a powerful multimodal model that can handle a wide variety of different tasks and data formats.

• vit-base-nsfw-detector↗

  AdamCodd

  This model is a fine-tuned version of vit-base-patch16-384 on around 25_000 images (drawings, photos...). It achieves the following results on the evaluation set: - Loss: 0.0937 - Accuracy: 0.9654 ** New [07/30] **: I created a new ViT model specifically to detect NSFW/SFW images for stable diffusion usage (read the disclaimer below for the reason): **AdamCodd/vit-nsfw-stable-diffusion**. **Disclaimer**: This model wasn't made with generative images in mind! There is no generated image in the dataset used here, and it performs significantly worse on generative images, which will require another ViT model specifically trained on generative images. Here are the model's actual scores for generative images to give you an idea: - Loss: 0.3682 (↑ 292.95%) - Accuracy: 0.8600 (↓ 10.91%) - F1: 0.8654 - AUC: 0.9376 (↓ 5.75%) - Precision: 0.8350 - Recall: 0.8980

• granite-3.0-8b-base↗

  ibm-granite

  **Model Summary:** Granite-3.0-8B-Base is a decoder-only language model to support a variety of text-to-text generation tasks. It is trained from scratch following a two-stage training strategy. In the first stage, it is trained on 10 trillion tokens sourced from diverse domains. During the second stage, it is further trained on 2 trillion tokens using a carefully curated mix of high-quality data, aiming to enhance its performance on specific tasks. **Supported Languages:** English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, and Chinese. Users may finetune Granite 3.0 models for languages beyond these 12 languages. **Intended use:** Prominent use cases of LLMs in text-to-text generation include summarization, text classification, extraction, question-answering, and more. All Granite Base models are able to handle these tasks as they were trained on a large amount of data from various domains. Moreover, they can serve as baseline to create specialized models for specific application scenarios.