전체 검색

AI 모델 2119

• Yanfei-v2-SamCool-i1-GGUF↗

  mradermacher

• Parable-Qwen3-8B-Claude-Fable-5-GGUF↗

  AnkitAI

  > **~6 GB of RAM is all you need.** Laptop, mid-range GPU, yesterday's desktop — the Q4 build runs > anywhere with that much headroom. One command and you have a private, offline reasoning model on your machine: > > **🔮 v2 is coming.** The 3B just got the v2 treatment (13× corpus, rebuilt recipe) — the same upgrade lands here next. Same links, in-place. **📦 Full family.** This 8B is the largest Parable, alongside Parable-Qwen3-4B — browse the full collection for every size, quant, and eval report.

• DaringLotus-SnowLotus-10.7b-IQ-GGUF↗

  BlueNipples

  The most recent version of llama.cpp has broken historical GGUFs, so I am uploading a few requants to preserve these two models compatibility. These will be called v3 in the file naming even though they are the same model. 3-4x Importance Matrix GGUFs and 3-4x regular GGUFs for https://huggingface.co/BlueNipples/SnowLotus-v2-10.7B and https://huggingface.co/BlueNipples/DaringLotus-v2-10.7b. I added a few more quants. I'm super happy with these merges, they turned out great. Basically Daring is the slightly more creative/prose oriented one, but also slightly less coherent. Daring basically nessesitates regens/swipes. They both have excellent prose for their size that is largely not very gpt-ish and are able to often take story context, lore entries and character card info into account. You can probably use these as your mainstay - which especially helpful if you GPU struggles with 13b, and honestly I think these models are _probably_ equal to or better than any 13b anyway. I might be wrong, but I do think they are very good compared to anything I've personally run. See the individual model cards for merge recipe details.

• CLIP-convnext_large_d_320.laion2B-s29B-b131K-ft-soup↗

  laion

  1. Model Details 2. Uses 3. Training Details 4. Evaluation 5. Acknowledgements 6. Citation A series of CLIP ConvNeXt-Large (w/ extra text depth, vision MLP head) models trained on the LAION-2B (english) subset of LAION-5B using OpenCLIP. The models utilize: * the timm ConvNeXt-Large model (`convnext_large`) as the image tower * a MLP (`fc - gelu - drop - fc`) head in vision tower instead of the single projection of other CLIP models * a text tower with same width but 4 layers more depth than ViT-L / RN50x16 models (depth 16, embed dim 768).

• leosamsHelloworldXL_helloworldXL70↗

  misri

• SilvermoonMix-Anima-Evolved↗

  silvermoong

  An Anima fine-tune by silvermoong, using the "Evolved" training strategy with a focus on stable hands, a neutral art style that works with artist tags, and multiple characters. **Update ComfyUI or Forge Neo to the latest version and restart before loading either 2.9B model.** Older versions may not load or run the model correctly. The standard and Turbo BF16 checkpoints are 5.84 GB each. The standard model also has a 3.16 GB mixed INT8 + ConvRot version, quantized from its BF16 weights (45.96% smaller). All variants use the same separate text encoder and VAE listed below. SHA-256 checksums are in releases.json.

• coreml-sam2.1-baseplus↗

  apple

• stable-diffusion-3.5-large-gguf↗

  city96

  As this is a quantized model not a finetune, all the same restrictions/original license terms still apply. Place model files in `ComfyUI/models/unet` - see the GitHub readme for further install instructions.

• Qwen3.8-27B-UD-Q4_K_XL-layers↗

  meshllm

  This package is derived from unsloth/Qwen3.8-27B-GGUF and keeps the original GGUF distribution split into per-layer artifacts for distributed inference. For upstream architecture details, chat template guidance, sampling recommendations, license terms, and benchmark notes, see the source model card: unsloth/Qwen3.8-27B-GGUF. Generated by the Mesh LLM HF Jobs splitter from `mesh-llm` ref `29bdf713d769a45cd299f4a4f7cea5333d700efd`. Each artifact is checksummed as it is written, uploaded to this repository, and removed from the job workspace before the next artifact is produced.

• ade20k_semantic_eomt_large_512↗

  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 🤗