サイト内検索
AI モデル 2119
- Qwen-Edit-2509-Upscale-LoRA↗
vafipas663
**V4 Candidates** LoRA A: qwen-edit-enhance-v4-16_a_000003000/1000, LoRA B: qwen-edit-enhance-v4-64_a_000003750/4750 Prompt: Enhance image quality, followed by textual description of the scene. The more descriptive it is, the better the upscale effect will be All images below are from 8 step Lighting LoRA. * ModelSamplingAuraFlow is a must, shift must be kept **below** 0.3. With higher resolutions, such as image 3, you can set it as low as 0.02 * Samplers - LCM (best), Euler_Ancestral, then Euler * Schedulers all work and give varying results in terms of smoothness
- LLaVA-NeXT-Video-7B-hf↗
llava-hf
Check out also the Google Colab demo to run Llava on a free-tier Google Colab instance: [](https://colab.research.google.com/drive/1CZggLHrjxMReG-FNOmqSOdi4z7NPq6SO?usp=sharing) Disclaimer: The team releasing LLaVa-NeXT-Video did not write a model card for this model so this model card has been written by the Hugging Face team. **Model type:** LLaVA-Next-Video is an open-source chatbot trained by fine-tuning LLM on multimodal instruction-following data. The model is buit on top of LLaVa-NeXT by tuning on a mix of video and image data to achieve better video understanding capabilities. The videos were sampled uniformly to be 32 frames per clip. The model is a current SOTA among open-source models on VideoMME bench. Base LLM: lmsys/vicuna-7b-v1.5
- LTXV-13B-0.9.8-distilled-GGUF↗
QuantStack
The model files can be used in ComfyUI with the ComfyUI-GGUF custom node. Place the required model(s) in the following folders: The old example workflow from last version should still work, youll just need to switch the vae and the gguf. **Example workflow** - based on the official example workflow *As this is a quantized model not a finetune, all the same restrictions/original license terms still apply.*
- Srijith-rkr_-_deepseek_base_1e-4_NO_cot_only_failed_samples_300_ckpt-gguf↗
RichardErkhov
- Qwen3.8-27B-nvfp4full-NInfer↗
cometkim
The repository contains a fuller-NVFP4 weight profile of Qwen/Qwen3.8-27B in the native NInfer `.ninfer` artifact format, with the z-lab/Qwen3.8-27B-DFlash2 block-diffusion speculative drafter embedded in the same image. It runs on NInfer **v3** engines — Text and MTP components through format/shape routes the upstream runtime already implements, the optional NVFP4-encoded DFlash2 companion through the fork's runtime support. It is not a Transformers checkpoint, Safetensors distribution, or GGUF file. Compared with the official `nvfp4` profile it extends NVFP4 from the layers 0–55 MLP to nearly the whole Text backbone: - every Text `mlp/gate_up` and `mlp/down` is NVFP4 (the layers 0–55 words are copied bit-exactly from unsloth/Qwen3.8-27B-NVFP4; layers 56–63 are quantized locally from the official BF16 checkpoint); - every GDN `query_key_value_z` is NVFP4, and GDN `output` is NVFP4 on 47 of 48 layers; - full-attention `query_key_gate_value` is NVFP4 on the ten deepest layers and `attention/output` on 14/16, with 9 BF16 exception parents retaining the registered Qwen3.6-27B NVFP4 exception pattern; - the token embedding and full output head use groupwise W8G32_F16S instead of row-scaled FP8. This yields 247 NVFP4 parents with 247 site-level input divisors. Locally quantized parents use the documented encoder profile `NVFP4_MAXABS_DIVISOR_RNE_V1` (per-tensor FP32 divisor, per-16 E4M3FN block scales, RNE E2M1 codes) with site divisors calibrated by streaming the official BF16 checkpoint. The complete contract, encoder, calibration corpus, and validation evidence are `docs/maintainer/qwen3.8-27b-artifact.md` §14 in the fork.
- RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-GGUF↗
deadbydawn101
> **The most comprehensive open-source security agent model — in GGUF.** Runs in Ollama, LM Studio, llama.cpp, vLLM, and any GGUF runtime. 51/51 LoRA tensors merged. Identical to the MLX version. **Both versions are identical** — same 51/51 LoRA tensors, same 745K+ training data, same 12 training rounds. **People are NOT getting the most out of local LLMs.** A 35B MoE at Q4_K_M gives dramatically better output than a 7B model at the SAME speed — because only 3B params activate per token.
- GLM-5.3-UNCENSORED-EXL3-3.0bpw↗
Infatoshi
EXL3 quantization of dealignai/GLM-5.3-UNCENSORED-FP8, itself a weight-edited (no fine-tune) variant of zai-org/GLM-5.3. The edit is documented in `CRACK_SURGERY.json` (copied unchanged from the source repo). This repo is not affiliated with dealignai or Z.ai. tau2-bench (airline, retail), Pass^1. Agent temperature 1.0, top_p 0.95; user simulator and judges GPT-4.1 at temperature 0. The reference is stock GLM-5.3 (not the uncensored edit) served at FP8 by Z.ai via OpenRouter, run through the same harness. The difference therefore mixes the dealign weight edit and this quantization. Per trial: airline baseline 0.740 / 0.680, quant 0.600 / 0.680; retail baseline 0.465 / 0.544, quant 0.482. ± is one binomial standard error. Neither difference is statistically significant at these sample sizes; treat the airline point estimate as a possible small regression rather than a measured one.
- sentence-t5-base↗
sentence-transformers
This is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space. The model works well for sentence similarity tasks, but doesn't perform that well for semantic search tasks. This model was converted from the Tensorflow model st5-base-1 to PyTorch. When using this model, have a look at the publication: Sentence-T5: Scalable sentence encoders from pre-trained text-to-text models. The tfhub model and this PyTorch model can produce slightly different embeddings, however, when run on the same benchmarks, they produce identical results. The model uses only the encoder from a T5-base model. The weights are stored in FP16.
- SmolLM3-3B-trellis-3inst-4bpw-kernel↗
xv0y5ncu
**GLQ (4.0 bits/weight)** quantization of `HuggingFaceTB/SmolLM3-3B` using a **trellis-coded codebook (QTIP TCQ) + randomized Hadamard transform (RHT) + LDLQ**. Weights stay compressed in memory and are decoded on the fly by a fused CUDA kernel. > **Where GLQ shines:** the size-vs-accuracy win is strongest at **2–4 bits/weight**. For even more headroom, combine it with the **E8 KV cache** (≈4× smaller KV cache → longer context in the same VRAM). > **Requires `glq >= 0.7.0`.** This checkpoint stores the trellis in the kernel > (MMA-fragment) layout that the lookup-free 3INST CUDA kernels consume > (`trellis_layout: "kernel"` in `config.json`); older glq versions cannot decode it.
- qwen-image-2.1-qma-camera-angles↗
gribbliest
Give it an image and a camera position. It renders the same subject, scene and style from the new angle. Orbit all the way around, raise or lower the camera, or pull in for a close-up. This is an edit LoRA for Qwen Image 2.1, so it always needs a source image. Directions are relative to the original camera, not the subject. "Right side view" means the camera swings 90 degrees to its own right around the subject. "Back view" takes it all the way around behind. Rather than typing prompts, you can drag a camera around to build them with the QMA Multi-Angle LoRA Node by CoreyCorza.