我的收藏提交工具
FOXAI SEARCH

全站搜索

AI 模型 2119

  • FLUX.1-schnell-GGUF↗

    unsloth

    `FLUX.1 [schnell]` is a 12 billion parameter rectified flow transformer capable of generating images from text descriptions. For more information, please read our blog post. 1. Cutting-edge output quality and competitive prompt following, matching the performance of closed source alternatives. 2. Trained using latent adversarial diffusion distillation, `FLUX.1 [schnell]` can generate high-quality images in only 1 to 4 steps. 3. Released under the `apache-2.0` licence, the model can be used for personal, scientific, and commercial purposes. We provide a reference implementation of `FLUX.1 [schnell]`, as well as sampling code, in a dedicated github repository. Developers and creatives looking to build on top of `FLUX.1 [schnell]` are encouraged to use this as a starting point. The FLUX.1 models are also available via API from the following sources - bfl.ml (currently `FLUX.1 [pro]`) - replicate.com - fal.ai - mystic.ai `FLUX.1 [schnell]` is also available in Comfy UI for local inference with a node-based workflow. To use `FLUX.1 [schnell]` with the 🧨 diffusers python library, first install or upgrade diffusers To learn more check out the diffusers documentation --- - This model is not intended or able to provide factual information. - As a statistical model this checkpoint might amplify existing societal biases. - The model may fail to generate output that matches the prompts. - Prompt following is heavily influenced by the prompting-style.

  • coreml-sam2.1-large↗

    apple

  • spot-wetfloor-convnext-tiny↗

    builtbyd3v

    Binary image classifier: `wet` (visible standing water or a wet surface) vs `dry`. Fine-tuned from facebook/convnext-tiny-224 (Apache-2.0). Built for SPOT Dispatch, a readiness board for Boston Dynamics Spot inspection routes. Spot's user guide says not to walk on wet or slippery surfaces, so the board can flag route photos that look wet for an operator to review. `onnx/model_quantized.onnx` (33.8 MB vs 111.4 MB fp32) loads with `{ dtype: "q8" }`. It is onnxruntime `quantize_dynamic`, uint8 weights, MatMul/Gemm layers only (convs stay fp32). Scored with onnxruntime on the same test split: q8 misses one more same-source wet image and changes 1 of 752 test predictions vs fp32 (kit images: same calls). Use fp32 when recall matters more than load time. Quantizing the convs too (28 MB) lost 2 same-source wet images, so it is not shipped.

  • mms-tts-bul↗

    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.

  • z-image-cyberrealistic-v9-gguf↗

    gimly13

    The key difference between Z-Image CyberRealistic v9 and v8 is not merely that it is "another version," but rather a shift in prompting philosophy. V9 moves away from tag stacks toward descriptive language: it better understands natural phrasing about lighting, scene and materials, relies less on booster words like "masterpiece" or "8k", and delivers more natural-looking skin, consistent anatomy and realistic lighting. Below are six prompts that demonstrate these differences in practice. All examples were generated with the same seed and sampler settings on both v8 and v9 for a fair comparison. 💡 Tip: These PNGs have the **ComfyUI workflow and prompt embedded!** Don't just copy the image from the browser preview. Go to the Files and versions tab, open the image there, and click "Download" to get the original file with all metadata intact.

  • bert_uncased_L-6_H-768_A-12↗

    Google

    This is the set of 24 BERT models referenced in Well-Read Students Learn Better: On the Importance of Pre-training Compact Models (English only, uncased, trained with WordPiece masking). We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher. Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity.

  • Z-Image-Turbo-Fun-Controlnet-Union-2.1↗

    alibaba-pai

    - **[2026.02.26]** Update to version 2602, with support for Gray Control. - **[2026.01.12]** Update to version 2601, with support for Scribble Control. Added lite models (1.9GB, 5 layers). Retrained Control and Tile models with enriched mask varieties, improved training schedules, and multi-resolution control images (512~1536) to fix mask pattern leakage and large `control_context_scale` artifacts. - **[2025.12.22]** Performed 8-step distillation on v2.1 to restore acceleration lost when applying ControlNet. Uploaded a tile model for super-resolution. - **[2025.12.17]** Fixed v2.0 typo (`control_layers` used instead of `control_noise_refiner`), which caused double forward pass and slow inference. Speed restored in v2.1. - This ControlNet is applied to 15 layer blocks and 2 refiner layer blocks (Lite models: 3 layer blocks and 2 refiner layer blocks). It supports multiple control conditions including Canny, HED, Depth, Pose, and MLSD (supporting Scribble in 2601 models and Gray in 2602 models). - Inpainting mode is also supported. For inpaint mode, use a larger `control_context_scale` for better image continuity. - **Training Process:** - **2.0:** Trained from scratch for 70,000 steps on 1M high-quality images (general and human-centric content) at 1328 resolution with BFloat16 precision, batch size 64, learning rate 2e-5, and text dropout ratio 0.10. - **2.1:** Continued training from 2.0 weights for 11,000 additional steps after fixing a typo, using the same parameters and dataset. - **2.1-8-steps:** Distilled from version 2.1 using an 8-step distillation algorithm for 5,500 steps. - **Note on Steps:** - **2.0 and 2.1:** Higher `control_context_scale` values may require more inference steps for better results, likely because the control model has not been distilled. - **2.1-8-steps:** Use 8 steps for inference. - Adjust `control_context_scale` (optimal range: 0.65–1.00) for stronger control and better detail preservation. A detailed prompt is highly recommended for stability. - In versions 2.0 and 2.1, applying ControlNet to Z-Image-Turbo caused loss of acceleration capability and blurry images. For strength and step count testing details, refer to Scale Test Results (generated with version 2.0). The old 8-steps model had bright spots/artifacts when the control_context_scale was too large, while the new version does not.

  • Samantha-v6.5-i1-GGUF↗

    mradermacher

  • Srijith-rkr_-_deepseek_base_1e-3_NO_cot_all_samples_100_ckpt-gguf↗

    RichardErkhov

  • decider-0.8b↗

    Mapika

    The smallest decider: a language model that does not generate text. It reads a **state** (a string or any JSON value) and a set of **typed questions** and returns a probability distribution for every question from a single forward pass: Choice (2-255 options, optionally described), Score (2-10 described levels), Noul (probability of yes). No decoding, no parsing, no output outside the options you defined. Same code, same wire format (`POST /v1/systemone`, TypeSafe Jev's format), same training recipe as the 2B: one epoch of `scripts/train.sh full` from `Qwen/Qwen3.5-0.8B-Base` over the full mixture (1.47M examples, 455M tokens, 4.5 h on one GH200). All six repositories share one interface (`decider.infer.Decider`, `POST /v1/systemone` in TypeSafe's format) and one readout: the letter logits at an answer slot, softmaxed over the options. Pick by size and input. Code, data registry, training scripts, the changelog and the per-version history: https://github.com/Mapika/decider.