我的收藏提交工具
FOXAI SEARCH

全站搜索

AI 模型 2119

  • mask2former-swin-tiny-coco-panoptic↗

    facebook

    Mask2Former model trained on COCO panoptic segmentation (tiny-sized version, Swin backbone). It was introduced in the paper Masked-attention Mask Transformer for Universal Image Segmentation and first released in this repository. Disclaimer: The team releasing Mask2Former did not write a model card for this model so this model card has been written by the Hugging Face team. Mask2Former addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation. Mask2Former outperforms the previous SOTA, MaskFormer both in terms of performance an efficiency by (i) replacing the pixel decoder with a more advanced multi-scale deformable attention Transformer, (ii) adopting a Transformer decoder with masked attention to boost performance without without introducing additional computation and (iii) improving training efficiency by calculating the loss on subsampled points instead of whole masks.

  • mdeberta-v3-base↗

    Microsoft

    DeBERTa 通过解耦注意力机制和增强的掩码解码器，对 BERT 和 RoBERTa 模型进行了改进。凭借这两项优化，在使用 80GB 训练数据的多数 NLU 任务上，DeBERTa 的性能均优于 RoBERTa。在 DeBERTa V3 中，我们进一步提升了 DeBERTa 的效率，采用了基于 ELECTRA 风格的预训练方法，并引入了梯度解耦嵌入共享机制。与 DeBERTa 相比，我们的 V3 版本显著提高了模型在下游任务上的表现。有关新模型的更多技术细节，请参阅我们的论文。mDeBERTa 是 DeBERTa 的多语言版本，其结构与 DeBERTa 相同，并使用 CC100 多语言数据进行训练。mDeBERTa V3 基础模型包含 12 层，隐藏层大小为 768。该模型具有 8600 万 backbone 参数，词汇表包含 25 万个 token，其中嵌入层引入了 1.9 亿个参数。该模型以 2.5T 的 CC100 数据进行训练，方式类似于 XLM-R。

  • qwen36-35b-a3b-colibri-i4-gs64↗

    Kreuzzelg

    colibri container for **Qwen3.6-35B-A3B**, quantized to **group-scaled int4** (one f32 scale per 64 input elements per row, `expert_gs=64`, ~22 GB). This is the higher-accuracy sibling of the per-row container `Kreuzzelg/qwen36-35b-a3b-colibri-i4`: in a controlled A/B against an int8 anchor it cuts the first-token logit error by **~44 %** (mean cosine 0.98777 → 0.99313 over 4 prompts). Same self-contained layout (bundled `tokenizer.json`, flat `config.json`). **Which one should I use?** The per-row container is the current PR default and runs on the plain `qwen36-engine` branch. This gs64 container needs the group-scaled expert path (`matmul_q_gs`), which lives on the `gs64-ab` branch (pending upstream). If you're on that branch and want the best quality, use this one; the container size grows ~10 % (20 → 22 GB) and decode speed is unchanged. Behavior: greedy 512-token generation on 4 reasoning prompts showed **no degenerate loops** on either container — the group scaling is a quality improvement here, not a fix for a runaway-generation bug.

  • TIPO-500M↗

    KBlueLeaf

    500M LLaMA arch model trained for TIPO. Tech Report: https://arxiv.org/abs/2411.08127 In this project, we introduce "TIPO" (**T**ext to **I**mage with text presampling for **P**rompt **O**ptimization), an innovative framework designed to significantly enhance the quality and usability of Text-to-Image (T2I) generative models. TIPO utilizes the Large Language Models (LLMs) to perform "Text Presampling" within the inference pipeline of text-to-image generative modeling. By refining and extending user input prompts, TIPO enables generative models to produce superior results with minimal user effort, making T2I systems more accessible and effective for a wider range of users. Use updated version of DTG extension (renamed to z-tipo-extension), current version of z-tipo-extension support stable-diffusion-webui, stable-diffusion-webui-forge and ComfyUI. SD-Next haven't been tested. https://github.com/KohakuBlueleaf/z-tipo-extension

  • llama-3-8b-samantha-GGUF↗

    mradermacher

  • gemma-4-26B-A4B-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.

  • Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-APEX-GGUF↗

    mudler

    ⚡ Each donation = another big MoE quantized I host 25+ free APEX MoE quantizations as independent research. My only local hardware is an NVIDIA DGX Spark (122 GB unified memory), enough for ~30-50B-class MoEs, but bigger ones (200B+) require rented compute on H100/H200/Blackwell, typically $20-100 per quant. If APEX quants are useful to you, your support directly funds those bigger runs. **APEX (Adaptive Precision for EXpert Models)** quantizations of hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled. APEX is a quantization strategy for Mixture-of-Experts (MoE) models. It classifies tensors by role (routed expert, shared expert, attention) and applies a layer-wise precision gradient — edge layers get higher precision, middle layers get more aggressive compression. I-variants use diverse imatrix calibration (chat, code, reasoning, tool-calling, agentic traces, Wikipedia).

  • coco_panoptic_eomt_large_640↗

    tue-mps

    **EoMT (Encoder-only Mask Transformer)** is a Vision Transformer (ViT) architecture designed for high-quality and efficient image segmentation. It was introduced in the CVPR 2025 highlight paper: **Your ViT is Secretly an Image Segmentation Model** by Tommie Kerssies, Niccolò Cavagnero, Alexander Hermans, Narges Norouzi, Giuseppe Averta, Bastian Leibe, Gijs Dubbelman, and Daan de Geus. > **Key Insight**: Given sufficient scale and pretraining, a plain ViT along with additional few params can perform segmentation without the need for task-specific decoders or pixel fusion modules. The same model backbone supports semantic, instance, and panoptic segmentation with different post-processing 🤗 The original implementation can be found in this repository.

  • Qwen3-Reranker-0.6B-GGUF-llama_cpp↗

    Voodisss

    Working GGUF of Qwen/Qwen3-Reranker-0.6B 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:** Q4_K_M (0.37 GB) is the sweet spot for 0.6B — 3x smaller than F16 with only 0.3% quality loss. Below Q4_K_M, quality starts to degrade: Q4_0 and Q3_K_M drop ~2%, and Q2_K is unusable (-28.7%). Smaller models are more sensitive to quantization than larger ones.

  • K2-Horizon-3.7B↗

    IFM

    K2-Horizon-3.7B is the small dense member of the K2-Horizon family: a 3.7B-core decoder-only model with a 512K context window. - **Strong small-model baseline.** A dense model evaluated on the same agentic, coding, and reasoning benchmarks as the rest of the family. - **512K context.** Native 524,288-token context from the midtraining stages onward. - **Intermediate checkpoints.** Intermediate checkpoints are released so capability changes can be studied across training rather than at a single checkpoint. - **Fully open.** Training data and recipe, training code, and evaluation resources are public. The chart at the top of this card shows K2-Horizon-3.7B against selected reference models. The table below lists every comparison model used in the figure.