全站搜索

AI 模型 2119

• mms-tts-lao↗

  facebook

  This model is part of Facebook's Massively Multilingual Speech project, aiming to provide speech technology across a diverse range of languages. You can find more details about the supported languages and their ISO 639-3 codes in the MMS Language Coverage Overview, and see all MMS-TTS checkpoints on the Hugging Face Hub: facebook/mms-tts. VITS (**V**ariational **I**nference with adversarial learning for end-to-end **T**ext-to-**S**peech) is an end-to-end speech synthesis model that predicts a speech waveform conditional on an input text sequence. It is a conditional variational autoencoder (VAE) comprised of a posterior encoder, decoder, and conditional prior. A set of spectrogram-based acoustic features are predicted by the flow-based module, which is formed of a Transformer-based text encoder and multiple coupling layers. The spectrogram is decoded using a stack of transposed convolutional layers, much in the same style as the HiFi-GAN vocoder. Motivated by the one-to-many nature of the TTS problem, where the same text input can be spoken in multiple ways, the model also includes a stochastic duration predictor, which allows the model to synthesise speech with different rhythms from the same input text.

• Qwen3.8-27B-MTP-NVFP4↗

  sakamakismile

  - **MTP draft head preserved in bf16** and wired into the index — speculative decoding works out of the box (`method: qwen3_5_mtp`). - Vision tower, `lm_head`, DeltaNet `conv1d`, and the MTP head are kept in bf16; everything else is NVFP4 W4A4. - Quantized **GPU-resident**: the full bf16 model dispatched across 8× RTX PRO 2000 Blackwell (16 GB) with `device_map=auto` — calibration itself took **68 seconds** (32 samples × 8192 tok). - Serves on SM120 (Blackwell) with vanilla vLLM v0.22.0 — compressed-tensors auto-detected, no `--quantization` flag. Single-stream prefill (prefix cache disabled, mean of 3): **3,820 tok/s @ 8k**, 3,299 @ 32k, 2,363 @ 100k prompt tokens. GPU KV cache at TP=4/128k: 1,025,977 tokens (7.83× concurrency). On boards without P2P add `NCCL_P2P_DISABLE=1` and `--disable-custom-all-reduce`.

• SamQuant-Qwen3.8-Flash-Next-Uncensored-MLX-Serve-Mix-C58↗

  LOKSUN-158

• german-sentiment-bert↗

  oliverguhr

  This model was trained for sentiment classification of German language texts. To achieve the best results all model inputs needs to be preprocessed with the same procedure, that was applied during the training. To simplify the usage of the model, we provide a Python package that bundles the code need for the preprocessing and inferencing. The model uses the Googles Bert architecture and was trained on 1.834 million German-language samples. The training data contains texts from various domains like Twitter, Facebook and movie, app and hotel reviews. You can find more information about the dataset and the training process in the paper. To get started install the package from pypi:

• sam3-litetext-s0↗

  vil-uob

  SAM3-LiteText was proposed in SAM3-LiteText: An Anatomical Study of the SAM3 Text Encoder for Efficient Vision-Language Segmentation by Chengxi Zeng, Yuxuan Jiang, Ge Gao, Shuai Wang, Duolikun Danier, Bin Zhu, Stevan Rudinac, David Bull, and Fan Zhang. SAM3-LiteText is a lightweight variant of SAM3 that replaces the heavy SAM3 text encoder (353M parameters) with a compact MobileCLIP-based text encoder optimized through knowledge distillation. The SAM3 ViT-H image encoder is kept intact. This reduces text encoder parameters by up to 88% while maintaining segmentation performance comparable to the original model. *Vision-language segmentation models such as SAM3 enable flexible, prompt-driven visual grounding, but inherit large, general-purpose text encoders originally designed for open-ended language understanding. In practice, segmentation prompts are short, structured, and semantically constrained, leading to substantial over-provisioning in text encoder capacity and persistent computational and memory overhead. In this paper, we perform a large-scale anatomical analysis of text prompting in vision-language segmentation, covering 404,796 real prompts across multiple benchmarks. Our analysis reveals severe redundancy: most context windows are underutilized, vocabulary usage is highly sparse, and text embeddings lie on low-dimensional manifold despite high-dimensional representations. Motivated by these findings, we propose SAM3-LiteText, a lightweight text encoding framework that replaces the original SAM3 text encoder with a compact MobileCLIP student that is optimized by knowledge distillation. Extensive experiments on image and video segmentation benchmarks show that SAM3-LiteText reduces text encoder parameters by up to 88%, substantially reducing static memory footprint, while maintaining segmentation performance comparable to the original model.*

• ThinkingCap-gemma-4-26B-A4B-it-GGUF↗

  bottlecapai

  **We recommend `ThinkingCap-gemma-4-26B-A4B-it-MXFP4_MOE.gguf` over `Q4_K_M` at this size**: it is the file evaluated below, and it held up better on our math benchmarks. Pull it with `-hf bottlecapai/ThinkingCap-gemma-4-26B-A4B-it-GGUF:MXFP4_MOE`. GGUF / llama.cpp quantizations of bottlecapai/ThinkingCap-gemma-4-26B-A4B-it — ThinkingCap's shorter-thinking post-training applied to gemma-4-26B-A4B-it, a mixture-of-experts model with image input. ➡️ Full model description, evaluation results, the measured thinking-token reduction and recommended sampling: see the main model card at bottlecapai/ThinkingCap-gemma-4-26B-A4B-it.

• bridgetower-large-itm-mlm-itc↗

  BridgeTower

  The BridgeTower model was proposed in "BridgeTower: Building Bridges Between Encoders in Vision-Language Representative Learning" by Xiao Xu, Chenfei Wu, Shachar Rosenman, Vasudev Lal, Wanxiang Che, Nan Duan. The model was pretrained on English language using masked language modeling (MLM) and image text matching (ITM)objectives. It was introduced in this paper and first released in this repository. The abstract from the paper is the following: Vision-Language (VL) models with the Two-Tower architecture have dominated visual-language representation learning in recent years. Current VL models either use lightweight uni-modal encoders and learn to extract, align and fuse both modalities simultaneously in a deep cross-modal encoder, or feed the last-layer uni-modal representations from the deep pre-trained uni-modal encoders into the top cross-modal encoder. Both approaches potentially restrict vision-language representation learning and limit model performance. In this paper, we propose BridgeTower, which introduces multiple bridge layers that build a connection between the top layers of uni-modal encoders and each layer of the cross-modal encoder. This enables effective bottom-up cross-modal alignment and fusion between visual and textual representations of different semantic levels of pre-trained uni-modal encoders in the cross-modal encoder. Pre-trained with only 4M images, BridgeTower achieves state-of-the-art performance on various downstream vision-language tasks. In particular, on the VQAv2 test-std set, BridgeTower achieves an accuracy of 78.73%, outperforming the previous state-of-the-art model METER by 1.09% with the same pre-training data and almost negligible additional parameters and computational costs. Notably, when further scaling the model, BridgeTower achieves an accuracy of 81.15%, surpassing models that are pre-trained on orders-of-magnitude larger datasets. Here is how to use this model to perform contrastive learning between image and text pairs:

• gemma-4-E2B-it-qat-q4_0-gguf↗

  Google

  > [!Note] > This model card is for the new versions of the Gemma 4 family optimized with Quantization-Aware Training (QAT), which allows preserving similar quality to bfloat16 while dramatically reducing the memory requirements to load the model. > Four versions of the QAT checkpoints are available: > * **Unquantized QAT checkpoints** (Q4_0): Half-precision weights extracted from the QAT pipeline, ideal for custom downstream compilation and research. Available for Gemma 4 E2B, E4B, 12B, 26B A4B, and 31B, and their drafter models. > * **GGUF** (Q4_0): Ready-to-deploy formats for broad ecosystem compatibility. Available for Gemma 4 E2B, E4B, 12B, 26B A4B, and 31B. > * **Mobile-optimized** (wNa8o8): A custom schema engineered explicitly for mobile hardware efficiency. It features targeted 2-bit decoding layers, optimized KV caches, and static activations to maximize VRAM savings. Available for Gemma 4 E2B and E4B. > * **Compressed Tensors** (w4a16): QAT checkpoints serialized in the compressed-tensors format for native, optimized inference with vLLM. Available for Gemma 4 E2B, E4B, 12B, and 31B. > * **Assistant Compatibility**: When using multi-token prediction (speculative decoding) with an assistant model alongside a QAT target model, the assistant model must also be a QAT checkpoint with the same precision to ensure compatibility. Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: **E2B**, **E4B**, **12B**, **26B A4B**, and **31B**. Their diverse sizes make them deployable in environments ranging from high-end phones to laptops and servers, democratizing access to state-of-the-art AI.

• GLM-5.3-Flash-Coder-RCO-152E-GGUF↗

  lapadatbogdan

  > **Update (October 29, 2026): Q3_K_M and Q4_K_M republished with an anti-loop fine-tune merged in.** A small > LoRA trained with verifiable-reward RL (GRPO with distillation from the full model) on multi-language coding tasks > removes the reasoning loops of the quantized files. Measured on Q3_K_M with the fine-tune applied as an adapter (40 held-out coding tasks x 2, and the Aider > "hmm" test, without any logit bias). The republished files contain the same fine-tune merged into the weights: > > | Q3_K_M | Before | After | > |---|---|---| > | Held-out coding tasks: looping runs | 11 of 80 | **0 of 80** | > | Held-out coding tasks: solved | 46 of 80 | **51 of 80** | > | Aider "hmm" test: exercises with a loop | 4 of 14 | **0 of 14** (14 of 14 solved) | > > The `--logit-bias` "hmm" ban below is no longer needed for these files. Q8_0 is unchanged: it does not show the > loop. The previous Q3_K_M / Q4_K_M files remain available in the repository history. > **New release (October 2026).** This repository used to hold REAP144, a 144-expert version calibrated on Python > agentic sessions only. REAP144 looped on general questions and on other programming languages and could get simple > arithmetic wrong. It is replaced by **RCO-152E**: 152 of 288 experts per layer, chosen by optimization on a mix of > agentic sessions, code in 21 languages, math and general text, then healed. The old REAP144 files were removed; they > remain available in the repository history. Recipe, mask, patches and evaluation harness: github.com/lapadatbogdan/GLM-5.3-Flash-Coder-RCO-152E

• Ltx2.3-VBVR-lora-I2V↗

  LiconStudio

  **To ensure training quality, we preprocessed the full 1,000,000 videos from the official dataset and randomly sample during training to maintain data diversity. We adopt the official parameters with batch_size=16 and rank=32 to prevent catastrophic forgetting caused by excessively large rank.** The VBVR dataset contains 100 reasoning task categories, with ~10,000 variants per task, totaling ~1M videos. Main task types include: This model is trained on the VBVR (Video Benchmark for Video Reasoning) dataset from video-reason.com.