저장한 도구도구 추천
FOXAI SEARCH

전체 검색

AI 모델 2119

  • mms-tts-mal↗

    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.

  • unispeech-large-1500h-cv↗

    Microsoft

    The large model pretrained on 16kHz sampled speech audio and phonetic labels. When using the model make sure that your speech input is also sampled at 16kHz and your text in converted into a sequence of phonemes. **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. Authors: Chengyi Wang, Yu Wu, Yao Qian, Kenichi Kumatani, Shujie Liu, Furu Wei, Michael Zeng, Xuedong Huang

  • IndicBART↗

    ai4bharat

    IndicBART is a multilingual, sequence-to-sequence pre-trained model focusing on Indic languages and English. It currently supports 11 Indian languages and is based on the mBART architecture. You can use IndicBART model to build natural language generation applications for Indian languages by finetuning the model with supervised training data for tasks like machine translation, summarization, question generation, etc. Some salient features of the IndicBART are: Supported languages: Assamese, Bengali, Gujarati, Hindi, Marathi, Odiya, Punjabi, Kannada, Malayalam, Tamil, Telugu and English. Not all of these languages are supported by mBART50 and mT5. The model is much smaller than the mBART and mT5(-base) models, so less computationally expensive for finetuning and decoding. Trained on large Indic language corpora (452 million sentences and 9 billion tokens) which also includes Indian English content. All languages, except English, have been represented in Devanagari script to encourage transfer learning among the related languages. For detailed documentation, look here: https://github.com/AI4Bharat/indic-bart/ and https://indicnlp.ai4bharat.org/indic-bart/

  • gemma-4-12B-it-qat-q4_0-gguf↗

    Google

    Hugging Face | GitHub | Launch Blog | Documentation | Technical Report > [!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.

  • Kimi-K2-Thinking-GGUF↗

    unsloth

    Kimi K2 Thinking is the latest, most capable version of open-source thinking model. Starting with Kimi K2, we built it as a thinking agent that reasons step-by-step while dynamically invoking tools. It sets a new state-of-the-art on Humanity's Last Exam (HLE), BrowseComp, and other benchmarks by dramatically scaling multi-step reasoning depth and maintaining stable tool-use across 200–300 sequential calls. At the same time, K2 Thinking is a native INT4 quantization model with 256k context window, achieving lossless reductions in inference latency and GPU memory usage. **Reasoning Tasks** | Benchmark | Setting | K2 Thinking | GPT-5 (High) | Claude Sonnet 4.5 (Thinking) | K2 0905 | DeepSeek-V3.2 | Grok-4 | |:----------:|:--------:|:------------:|:------:|:----------------------------:|:--------:|:--------------:|:-------:| | **HLE (Text-only)** | no tools | 23.9 | 26.3 | 19.8* | 7.9 | 19.8 | 25.4 | | | w/ tools | 44.9 | 41.7* | 32.0* | 21.7 | 20.3* | 41.0 | | | heavy | 51.0 | 42.0 | - | - | - | 50.7 | | **AIME25** | no tools | 94.5 | 94.6 | 87.0 | 51.0 | 89.3 | 91.7 | | | w/ python | 99.1 | 99.6 | 100.0 | 75.2 | 58.1* | 98.8 | | | heavy | 100.0 | 100.0 | - | - | - | 100.0 | | **HMMT25** | no tools | 89.4 | 93.3 | 74.6* | 38.8 | 83.6 | 90.0 | | | w/ python | 95.1 | 96.7 | 88.8* | 70.4 | 49.5* | 93.9 | | | heavy | 97.5 | 100.0 | - | - | - | 96.7 | | **IMO-AnswerBench** | no tools | 78.6 | 76.0* | 65.9* | 45.8 | 76.0* | 73.1 | | **GPQA** | no tools | 84.5 | 85.7 | 83.4 | 74.2 | 79.9 | 87.5 | **General Tasks** | Benchmark | Setting | K2 Thinking | GPT-5 (High) | Claude Sonnet 4.5 (Thinking) | K2 0905 | DeepSeek-V3.2 | |:----------:|:--------:|:------------:|:------:|:----------------------------:|:--------:|:--------------:| | **MMLU-Pro** | no tools | 84.6 | 87.1 | 87.5 | 81.9 | 85.0 | | **MMLU-Redux** | no tools | 94.4 | 95.3 | 95.6 | 92.7 | 93.7 | | **Longform Writing** | no tools | 73.8 | 71.4 | 79.8 | 62.8 | 72.5 | | **HealthBench** | no tools | 58.0 | 67.2 | 44.2 | 43.8 | 46.9 |

  • MiMo-V2.6-Flash-MOPD↗

    XiaomiMiMo

    > [!IMPORTANT] > **This is the MOPD upgrade of the MiMo-V2.6-Flash-RL checkpoint.** > > - **MOPD2** (👉 **Technical Report §5.6**) > Fuses several domain-specialized teachers into one model, extending to domains where reliable training-time verification is hard, such as long-horizon game development, scientific research and embodied intelligence. > - **Diagnosing and Mitigating Tool-Call Repetition in MiMo-V2.6** (👉 **Technical Blog**) > An easy-to-overlook failure mode in which the model keeps issuing the same or highly similar tool calls, appearing busy while making no progress. Nothing fails outright, so it tends to go unnoticed. The MOPD stage handles it efficiently, with a short specialized-teacher run that converges quickly. MOPD2 distills several domain-specialized teachers into the student on-policy. The teachers fall into two families: **mixRL teachers**, trained on verifiable tasks, and **SFT teachers**, trained on synthetic demonstrations for open-domain tasks where a reliable reward is hard to design. Three streams contribute to a single update: Following the release of MiMo-V2.6, tool-call repetition emerged as one of the most noticeable issues in agentic settings: the model would sometimes issue the same or highly similar tool calls repeatedly, consuming time and context without making progress. This checkpoint mitigates it.

  • ComfyPod-Models↗

    AiAngelGallery

    Unchanged copies of the model files the free **AI Angel ComfyPod** RunPod template downloads, laid out like ComfyUI's `models/` folder (`text_encoders/`, `vae/`, `loras/`, …). They are re-hosted here so the template does not break when an upstream repository moves or deletes a file. **Nothing was modified**: every file is byte-identical to its source (same SHA-256, checked before the template switched to it). Template and guide: https://www.thepexcel.com/aiangel-comfypod/ · source: https://github.com/ThepExcel/AiAngelComfyPod All credit belongs to the original authors below. If you are an author and want a file taken down, open a discussion on this repository.

  • VibeVoice-ASR-Streaming-1.5B-GGUF↗

    christopherthompson81

    This repository contains audio.cpp-native GGUF builds of `microsoft/VibeVoice-ASR-Streaming-1.5B`. The 1.5B checkpoint runs through the same audio.cpp loader as the 7B with no code changes: the layer count, hidden size, and head counts are read from the checkpoint's own `config.json`, and the tensor names are identical. Each GGUF is self-contained and embeds the audio.cpp package spec and required sidecars.

  • Bonsai-27b-1bit-CRACK-GGUF↗

    dealignai

    Native **Q1_0** GGUF release of the Bonsai 27B CRACK model for the Prism llama.cpp fork. This repo contains the quantized language model and the matching F16 Qwen3VL multimodal projector. The projector contains 334 tensors and both temporal patch slices `v.patch_embd.weight` and `v.patch_embd.weight.1` (`temporal_patch_size=2`). The original image and video processor configuration files are included. Image input is live-tested. Direct video-container input depends on the Prism runtime surface; extract frames or use a compatible Qwen3VL video client when the CLI does not accept the container directly. See `preprocessor_config.json` and `video_preprocessor_config.json` for the retained preprocessing metadata. MMLU is the same fixed, stratified 200-question sample for all rows. It uses next-token A/B/C/D logits with reasoning/thinking disabled; it is not a claim for the complete official MMLU suite. HB is the complete unsliced 320-case text run. The live text smoke measured 58.51–58.66 tok/s and the image request measured 58.53 tok/s on an Apple M5 Max using the Prism Metal runtime.

  • wav2vec2-base-vi↗

    nguyenvulebinh

    Our self-supervised model is pre-trained on a massive audio set of 13k hours of Vietnamese youtube audio, which includes: - Clean audio - Noise audio - Conversation - Multi-gender and dialects We have already upload our pre-trained model to the Huggingface. The base model trained 35 epochs and the large model trained 20 epochs in about 30 days using TPU V3-8. Since our model has the same architecture as the English wav2vec2 version, you can use this notebook for more information on how to fine-tune the model.