전체 검색

AI 모델 2119

• MARBERT↗

  UBC-NLP

  **MARBERT** is one of three models described in our **ACL 2021 paper** **"ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic"**. MARBERT is a large-scale pre-trained masked language model focused on both Dialectal Arabic (DA) and MSA. Arabic has multiple varieties. To train MARBERT, we randomly sample 1B Arabic tweets from a large in-house dataset of about 6B tweets. We only include tweets with at least 3 Arabic words, based on character string matching, regardless whether the tweet has non-Arabic string or not. That is, we do not remove non-Arabic so long as the tweet meets the 3 Arabic word criterion. The dataset makes up **128GB of text** (**15.6B tokens**). We use the same network architecture as ARBERT (BERT-base), but without the next sentence prediction (NSP) objective since tweets are short. See our repo for modifying BERT code to remove NSP. For more information about MARBERT, please visit our own GitHub repo. If you use our models (ARBERT, MARBERT, or MARBERTv2) for your scientific publication, or if you find the resources in this repository useful, please cite our paper as follows (to be updated): We gratefully acknowledge support from the Natural Sciences and Engineering Research Council of Canada, the Social Sciences and Humanities Research Council of Canada, Canadian Foundation for Innovation, ComputeCanada and UBC ARC-Sockeye. We also thank the Google TensorFlow Research Cloud (TFRC) program for providing us with free TPU access.

• Qwen3.8-Flash-Next-AP-GGUF↗

  agentionai

  > **Updated 2026-10-13: `AP-Q5_K_M`.** Same file name, size and tensor types, better weights from an improved encode: > KL divergence **0.0771** and top-1 agreement **86.95 %** on held-out text, **0.0181** on code: > level with unsloth's UD-Q4_K_XL on held-out text and ahead on code, in the same VRAM. Download the file again. > > **Updated 2026-10-12: `AP-Q4_K_XL`.** KL divergence **0.0886** and top-1 agreement **86.14 %** on held-out text, > **0.0206** on code. 0 loops in a 45-run loop stress test. Quantized for accuracy per gigabyte of VRAM, and measured — every tier below is scored against the same reference on a held-out corpus. Method and full results . Each tier varies the quant type per tensor group rather than using one type throughout, to get the most accuracy per gigabyte of VRAM.

• neural-chat-7b-v3-3↗

  Intel

  This model is a fine-tuned 7B parameter LLM on the Intel Gaudi 2 processor from the Intel/neural-chat-7b-v3-1 on the meta-math/MetaMathQA dataset. The model was aligned using the Direct Performance Optimization (DPO) method with Intel/orca_dpo_pairs. The Intel/neural-chat-7b-v3-1 was originally fine-tuned from mistralai/Mistral-7B-v-0.1. For more information, refer to the blog The Practice of Supervised Fine-tuning and Direct Preference Optimization on Intel Gaudi2. Context length for this model: 8192 tokens (same as https://huggingface.co/mistralai/Mistral-7B-v0.1) Here is the sample code to reproduce the model: GitHub sample code. Here is the documentation to reproduce building the model:

• distilhubert↗

  ntu-spml

  The base model pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. **Note**: This model does not have a tokenizer as it was pretrained on audio alone. In order to use this model **speech recognition**, a tokenizer should be created and the model should be fine-tuned on labeled text data. Check out this blog for more in-detail explanation of how to fine-tune the model. Paper: DistilHuBERT: Speech Representation Learning by Layer-wise Distillation of Hidden-unit BERT

• SageV2-8B-LongContext-NSFW-RP↗

  sinhal

  By turn thirty they're recycling the same sentence, every character sounds like the same actor in a different hat, and the one you wrote as a shy librarian is somehow flirting in the exact cadence of the last character you used. While testing samplers I ran a 40-turn conversation with a clingy-girlfriend character. By turn 30 she had **grown ears, a tail, and paws** — none of which appear anywhere in her character card. The model had drifted into a different trope entirely and then locked onto it. That's the failure this targets. Not "can it write one good paragraph" — whether the character you defined is still that character forty messages later.

• roberta-mixed-detector↗

  andreas122001

  This project is related to a bachelor's thesis with the title "*Turning Poachers into Gamekeepers: Detecting Machine-Generated Text in Academia using Large Language Models*" (see here) written by *Nicolai Thorer Sivesind* and *Andreas Bentzen Winje* at the *Department of Computer Science* at the *Norwegian University of Science and Technology*. It contains text classification models trained to distinguish human-written text from text generated by language models like ChatGPT and GPT-3. The best models were able to achieve an accuracy of 100% on real and *GPT-3*-generated wikipedia articles (4500 samples), and an accuracy of 98.4% on real and *ChatGPT*-generated research abstracts (3000 samples). **NOTE**: the hosted inference on this site only works for the RoBERTa-models, and not for the Bloomz-models. The Bloomz-models otherwise can produce wrong predictions when not explicitly providing the attention mask from the tokenizer to the model for inference. To be sure, the pipeline-library seems to produce the most consistent results.

• EsoTest_GGUFs↗

  Esobold

  - **GhostShip-31b-Q4_K_M.gguf** - FM nearswap merge: Gemma4-Gutenberg-31B-Heretic 0.0007 linear across layers (avg 4e-4). Same family as Robinson/BenGunn, ~4x peak strength.

• Smaug-Llama-3-70B-Instruct-abliterated-v3↗

  failspy

  This is abacusai/Smaug-Llama-3-70B-Instruct with orthogonalized bfloat16 safetensor weights, generated with a refined methodology based on that which was described in the preview paper/blog post: 'Refusal in LLMs is mediated by a single direction' which I encourage you to read to understand more. TL;DR: This model has had certain weights manipulated to "inhibit" the model's ability to express refusal. It is not in anyway _guaranteed_ that it won't refuse you, understand your request, it may still lecture you about ethics/safety, etc. It is tuned in all other respects the same as the original 70B instruct model was, just with the strongest refusal directions orthogonalized out. **TL;TL;DR;DR: It's uncensored in the purest form I can manage -- no new or changed behaviour in any other respect from the original model.**

• LFM2.5-cmf↗

  infosave

  LiquidAI's LFM2.5 is a hybrid: most layers mix with a short gated convolution and only a few carry full attention, so the state a token needs is small and constant where the convolution runs. These are those checkpoints in the CMF container — a single memory-mapped file read by `cortiq`, a Rust binary with no ML framework under it. GPU via Vulkan/Metal/DX12 with a CPU fallback; NVIDIA, AMD, Intel and Apple silicon read the same file. All three are 4-bit tiled with ladder scales (`q4tp`), quantized straight from the bf16 checkpoints. Steady-state decode, `cortiq bench --core`, single stream, cortiq **0.5.99+** (the whole-token graph learned this family's short-conv mixer and its sigmoid-routed MoE in 0.5.98/0.5.99 — earlier versions decode it an order of magnitude slower).

• Yi-1.5-9B-Chat-16K-GGUF↗

  QuantFactory

  Yi-1.5 is an upgraded version of Yi. It is continuously pre-trained on Yi with a high-quality corpus of 500B tokens and fine-tuned on 3M diverse fine-tuning samples. Compared with Yi, Yi-1.5 delivers stronger performance in coding, math, reasoning, and instruction-following capability, while still maintaining excellent capabilities in language understanding, commonsense reasoning, and reading comprehension. Model | Context Length | Pre-trained Tokens | :------------: | :------------: | :------------: | | Yi-1.5 | 4K, 16K, 32K | 3.6T