我的收藏提交工具
FOXAI SEARCH

全站搜索

AI 模型 2119

  • resnet50.tv2_in1k↗

    timm

    This model features: * ReLU activations * single layer 7x7 convolution with pooling * 1x1 convolution shortcut downsample

  • Yue2-3B-GGUF↗

    audio-cpp

    This repository contains audio.cpp-native GGUF weights for Yue2-3B. The model has been merged into the main branch. The demos below were generated with `yue2-3b-bf16.gguf` and `yue2-vae-f32.gguf`. The `_q8.wav` comparison files were generated with `yue2-3b-q8_0.gguf` and `yue2-vae-f16.gguf` using the same prompts and seeds. The `_q4_0.wav` comparison files were generated with `yue2-3b-q4_0.gguf` and `yue2-vae-f16.gguf`. Replace ` ` with your audio.cpp checkout and ` ` with this downloaded GGUF repo directory. To generate the Q4_0 versions, use the same commands and replace `--session-option yue2.model_gguf=yue2-3b-bf16.gguf` with `--session-option yue2.model_gguf=yue2-3b-q4_0.gguf`, and use `--session-option yue2.vae_gguf=yue2-vae-f16.gguf`. Measured with audio.cpp server mode on an RTX 5090, using the official `tonight-awake` longform example. Each run restarted the server and used one short warmup request before the measured longform request. Peak VRAM was sampled continuously during the measured request.

  • Codestral-22B-v0.1↗

    Mistral AI

    > [!TIP] > PRs to correct the `transformers` tokenizer so that it gives 1-to-1 the same results as the `mistral_common` reference implementation are very welcome! Codestral-22B-v0.1 is trained on a diverse dataset of 80+ programming languages, including the most popular ones, such as Python, Java, C, C++, JavaScript, and Bash (more details in the Blogpost). The model can be queried: - As instruct, for instance to answer any questions about a code snippet (write documentation, explain, factorize) or to generate code following specific indications - As Fill in the Middle (FIM), to predict the middle tokens between a prefix and a suffix (very useful for software development add-ons like in VS Code) After installing `mistral_inference`, a `mistral-chat` CLI command should be available in your environment.

  • resnet18d.ra2_in1k↗

    timm

    This model features: * ReLU activations * 3-layer stem of 3x3 convolutions with pooling * 2x2 average pool + 1x1 convolution shortcut downsample Recipe details: * RandAugment `RA2` recipe. Inspired by and evolved from EfficientNet RandAugment recipes. Published as `B` recipe in ResNet Strikes Back. * RMSProp (TF 1.0 behaviour) optimizer, EMA weight averaging * Step (exponential decay w/ staircase) LR schedule with warmup - **Model Type:** Image classification / feature backbone - **Model Stats:** - Params (M): 11.7 - GMACs: 2.1 - Activations (M): 3.3 - Image size: train = 224 x 224, test = 288 x 288 - **Papers:** - ResNet strikes back: An improved training procedure in timm: https://arxiv.org/abs/2110.00476 - Deep Residual Learning for Image Recognition: https://arxiv.org/abs/1512.03385 - Bag of Tricks for Image Classification with Convolutional Neural Networks: https://arxiv.org/abs/1812.01187 - **Original:** https://github.com/huggingface/pytorch-image-models

  • gemma-3-12b-it-bnb-4bit↗

    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 and 1B with 2 trillion tokens. 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.

  • Qwen3-Reranker-4B-GGUF-llama_cpp↗

    Voodisss

    Working GGUF of Qwen/Qwen3-Reranker-4B for llama.cpp. Converted 2025-03-09 with the official `convert_hf_to_gguf.py`. Benchmarked on MTEB AskUbuntuDupQuestions (361 queries) via llama-server `/v1/rerank` on RTX 3090. All quants produced from the same F16 source using `llama-quantize`. **Takeaway:** All quants from Q8_0 down to Q3_K_M are within ±1% of F16 — pick based on your VRAM budget. Q4_K_M (2.33 GB) is the sweet spot: 3.2x smaller than F16 with no measurable quality loss. **Avoid Q2_K** — it's the only quant with real degradation.

  • Qwen3.8-27B-MTPLX-Optimized-Quality↗

    Youssofal

    **MTPLX: the fastest way to run Qwen 3.8 on a Mac. Native multi-token-prediction speculative decoding on Apple Silicon, two to three times the speed of plain decoding, exact at any temperature.** The closest of the three MTPLX Qwen 3.8 builds to the original bf16 model. Every weight matrix at 8-bit, sensitive parts at 16-bit, native multi-token-prediction head kept, so MTPLX still drafts ahead and verifies in one pass. Pick this when you want the answer the full model would give and still want it fast. This is a Qwen3.8-27B MLX pack for MTPLX, the fastest way to run Qwen 3.8 27B on a Mac. All numbers on a MacBook Pro M5 Max, fans verified at maximum, sampled at the model's own settings. Conditions and sources: mtplx.com/benchmarks.

  • Arabic-SBERT-100K↗

    akhooli

    This is a sentence-transformers model finetuned from aubmindlab/bert-base-arabertv02. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. This model is trained on 100K samples filtered from the akhooli/arabic-triplets-1m-curated-sims-len dataset with 75K training and 25K validation. Trained for 5 epochs, with final training loss of 0.133 (using MatryoshkaLoss).

  • Qwen3.5-9B-Heretic-GGUF↗

    darrellbest

    GGUF builds of darrellbest/Qwen3.5-9B-Heretic. Qwen/Qwen3.5-9B with its refusal behaviour removed by Heretic using full-weight Arbitrary-Rank Ablation: **5/100 refusals** (original: 100/100) at **KL divergence 0.0403**. See the main repository for how it was made and measured. Converted from the bf16 safetensors with llama.cpp's `convert_hf_to_gguf.py` and quantized with `llama-quantize` (no imatrix). The mmproj file carries the vision encoder; load it alongside any of the three for image input. Every file was loaded in `llama-server` with the mmproj and in Ollama. All three answered ordinary prompts correctly, described a test image (a red circle and a blue square) correctly, and in Ollama answered 17 x 23 = 391 with thinking off. Thinking-mode reasoning, 4 arithmetic and word problems x 10 seeds with Qwen's recommended sampling: BF16 39/40, Q8_0 37/40, Q4_K_M 35/40 finished and correct (the original model in vLLM: 40/40); the misses ran out of the 4,000-token budget rather than answering wrongly.

  • FLUX.1-dev↗

    rayonlabs

    `FLUX.1 [dev]` 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, second only to our state-of-the-art model `FLUX.1 [pro]`. 2. Competitive prompt following, matching the performance of closed source alternatives . 3. Trained using guidance distillation, making `FLUX.1 [dev]` more efficient. 4. Open weights to drive new scientific research, and empower artists to develop innovative workflows. 5. Generated outputs can be used for personal, scientific, and commercial purposes as described in the [`FLUX.1 [dev]` Non-Commercial License](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md). We provide a reference implementation of `FLUX.1 [dev]`, as well as sampling code, in a dedicated github repository. Developers and creatives looking to build on top of `FLUX.1 [dev]` are encouraged to use this as a starting point.