Hugging Face Models (Daily)

A **7.3M parameter** T5-based model that answers general knowledge questions. Smaller than a mosquito's brain, but surprisingly capable!



Open: https://huggingface.co/ag14850/Mosquito

This model is a fine-tuned version of indolem/indobert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:



Open: https://huggingface.co/ahmatfauzy/alodokter-sentimen-id

***



Open: https://huggingface.co/asigalov61/Karaoke-Lyrics-Qwen3-0.6B

Using llama.cpp release b7779 for quantization.



Open: https://huggingface.co/bartowski/zai-org_GLM-4.7-Flash-GGUF

This is a BF16 model based on HeartCodec-oss https://huggingface.co/HeartMuLa/HeartCodec-oss



Open: https://huggingface.co/benjiaiplayground/HeartCodec-oss-bf16

This is a BF16 model based on HeartMuLa-oss-3B https://huggingface.co/HeartMuLa/HeartMuLa-oss-3B



Open: https://huggingface.co/benjiaiplayground/HeartMuLa-oss-3B-bf16

 



Open: https://huggingface.co/briaai/Fibo-Edit

Instruction-tuned **Qwen3-1.7B-Base** using **SFT (QLoRA/LoRA)** on **MBZUAI/LaMini-instruction**, then **merged** into a single model checkpoint for easy deployment (single-turn Q/A).



Open: https://huggingface.co/ericoh929/qwen3-1.7b-lamini-qlora-instruction-tuned

Ex0bit/GLM-4.7-Flash-PRISM



Open: https://huggingface.co/Ex0bit/GLM-4.7-Flash-PRISM

This is a merge of pre-trained language models created using mergekit.



Open: https://huggingface.co/gss1147/Darwin-Astral-4B-Coder

This is a **hybrid quantized version** of mPLUG/UI-S1-7B optimized for efficient GUI automation on consumer hardware.



Open: https://huggingface.co/Hadidiz9/UI-S1-7B-Hybrid-W4-Quanto

This model is a mixed int4 model with group_size 128 and symmetric quantization of meituan-longcat/LongCat-Flash-Thinking-2601 generated by intel/auto-round. Please follow the license of the original model.



Open: https://huggingface.co/INC4AI/LongCat-Flash-Thinking-2601-int4-mixed-AutoRound

This is an FP8 quantized version of Qwen3-VL-8B-Instruct, a powerful vision-language model for multimodal understanding and generation tasks.



Open: https://huggingface.co/JEILDLWLRMA/Qwen3VL-8B-Instruct-FP8

Open: https://huggingface.co/karham/karham

A production-ready PyTorch implementation of **1.58-bit ternary linear layers** that achieves **~19x memory compression** while maintaining high accuracy. Drop-in replacement for nn.Linear with optimized C++/CUDA kernels.



Open: https://huggingface.co/krisaujla/BitLinear

Open: https://huggingface.co/kwonhan/GLM-4.7-Flash-mixed-4-6-g32-mlx

Try LFM • Documentation • LEAP



Open: https://huggingface.co/LiquidAI/LFM2.5-1.2B-Thinking

- **Developed by:** Medical Swinburne University of Technology AI Team - **Funded by:** Swinburne University of Technology - **Language(s):** English - **License:** Apache 2.0



Open: https://huggingface.co/MedSwin/MedSwin-KD-SFT-PubMed-l

- **Developed by:** Medical Swinburne University of Technology AI Team - **Funded by:** Swinburne University of Technology - **Language(s):** English - **License:** Apache 2.0



Open: https://huggingface.co/MedSwin/MedSwin-KD-SFT-PubMed-map

> [!IMPORTANT] > Original Model Link : https://huggingface.co/skt/A.X-4.0-Light >



Open: https://huggingface.co/mlx-community/A.X-4.0-Light-MLX-BF16

> [!IMPORTANT] > Original Model Link : https://huggingface.co/Freepik/flux.1-lite-8B >



Open: https://huggingface.co/mlx-community/Flux-1.lite-8B-MLX-Q4

> [!IMPORTANT] > Original Model Link : https://huggingface.co/shuttleai/shuttle-3.1-aesthetic >



Open: https://huggingface.co/mlx-community/shuttle-3.1-aesthetic-MLX

> [!IMPORTANT] > Original Model Link : https://huggingface.co/shuttleai/shuttle-3.1-aesthetic >



Open: https://huggingface.co/mlx-community/shuttle-3.1-aesthetic-MLX-Q8

> [!IMPORTANT] > Original Model Link : https://huggingface.co/shuttleai/shuttle-jaguar >



Open: https://huggingface.co/mlx-community/shuttle-jaguar-MLX

GemmaX2-28-2B-v0.1 is an LLM-based translation model. It has been fintuned on GemmaX2-28-2B-Pretrain, which is a language model developed through continual pretraining of Gemma2-2B using a mix of 56 billion tokens from both monolingual and parallel data across 28 different languages. Please find more details in our paper: Multilingual Machine Translation wi…



Open: https://huggingface.co/ModelSpace/GemmaX2-28-2B-v0.1

static quants of https://huggingface.co/MuXodious/gemma-3n-E4B-it-absolute-heresy-MPOA



Open: https://huggingface.co/mradermacher/gemma-3n-E4B-it-absolute-heresy-MPOA-GGUF

weighted/imatrix quants of https://huggingface.co/MuXodious/gemma-3n-E4B-it-absolute-heresy-MPOA



Open: https://huggingface.co/mradermacher/gemma-3n-E4B-it-absolute-heresy-MPOA-i1-GGUF

This is a **GLM-4.7-Flash** fine-tune, produced at the request of McG-221 through P-E-W's Heretic (v1.1.0) abliteration engine merged with the Magnitude-Preserving Orthogonal Ablation PR.



Open: https://huggingface.co/MuXodious/GLM-4.7-Flash-impotent-heresy

A **Romansh dialect classifier** that predicts one of 6 varieties:



Open: https://huggingface.co/nomichaelno/romansh-classifier-v1

!



Open: https://huggingface.co/piotr-ai/Polanka_VL_v0.1_Qwen3_VL_4b_260120

!



Open: https://huggingface.co/piotr-ai/Polanka_VL_v0.1_Qwen3_VL_4b_260120_gguf

Diffusers-compatible transformer weights extracted from Phr00t/Qwen-Image-Edit-Rapid-AIO-NSFW-V20 for 4-step accelerated Qwen Image Edit inference.



Open: https://huggingface.co/prithivMLmods/Qwen-Image-Edit-Rapid-AIO-V20

TM-Vec 2s: A distilled Bi-LSTM for remote homology detection. 2s was trained using a teacher-student architecture.



Open: https://huggingface.co/scikit-bio/TMVec-2s

This is a sentence-transformers model finetuned from microsoft/codebert-base. 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.



Open: https://huggingface.co/shubharuidas/codebert-embed-base-dense-retriever

This is a model for High-definition magnification of the picture, trained on ``black-forest-labs/FLUX.2-klein-base-9B`, and it is mainly used for losslessly enlarging images to approximately 2K size.For use in `ComfyUI``.



Open: https://huggingface.co/starsfriday/FLUX.2-klein-AC-Style-LORA

To reduce looping and improve outputs, add --dry-multiplier 1.1. This is **not** the same as a repeat penalty.If dry-multiplier isn’t available, **disable repeat penalty** instead.



Open: https://huggingface.co/unsloth/GLM-4.7-Flash-GGUF

**AILO** (Artificial Intelligence Language Operator) is a 152M parameter Transformer-based language model trained from scratch.



Open: https://huggingface.co/xxrickyxx/ailo-152m

📐 Front-Right Isometric View LoRA 右前 45° 的等轴侧视角转换lora



Open: https://huggingface.co/zealman/front-right-isometric-view_lora_v1

**A.X Encoder** (pronounced "A dot X") is SKT's document understanding model optimized for Korean-language understanding and enterprise deployment. This lightweight encoder was developed entirely in-house by SKT, encompassing model architecture, data curation, and training, all carried out on SKT’s proprietary supercomputing infrastructure, TITAN. This mo…

Daily delta: +31,040 downloads / +2 likes

Open: https://huggingface.co/skt/A.X-Encoder-base

MapAnything is a simple, end-to-end trained transformer model that directly regresses the factored metric 3D geometry of a scene given various types of modalities as inputs. A single feed-forward model supports over 12 different 3D reconstruction tasks including multi-image sfm, multi-view stereo, monocular metric depth estimation, registration, depth compl…

Daily delta: +21,126 downloads / +4 likes

Open: https://huggingface.co/facebook/map-anything-apache

This is a GGUF quantized version of LTX-2. unsloth/LTX-2-GGUF uses Unsloth Dynamic 2.0 methodology for SOTA performance. - Important layers are upcasted to higher precision. - To use the model, read our guides for ComfyUI or stable-diffusion.cpp. - Uses tooling from ComfyUI-GGUF by city96.

Daily delta: +16,828 downloads / +55 likes

Open: https://huggingface.co/unsloth/LTX-2-GGUF

👋 Join our Discord community. 📖 Check out the GLM-4.7 technical blog, technical report(GLM-4.5). 📍 Use GLM-4.7-Flash API services on Z.ai API Platform. 👉 One click to GLM-4.7.

Daily delta: +15,168 downloads / +470 likes

Open: https://huggingface.co/zai-org/GLM-4.7-Flash

👋 Join our Discord community. 📖 Check out the GLM-4.7 technical blog, technical report(GLM-4.5). 📍 Use GLM-4.7-Flash API services on Z.ai API Platform. 👉 One click to GLM-4.7.

Daily delta: +4,387 downloads / +4 likes

Open: https://huggingface.co/cyankiwi/GLM-4.7-Flash-AWQ-4bit

STEP3-VL-10B

Daily delta: +2,913 downloads / +31 likes

Open: https://huggingface.co/stepfun-ai/Step3-VL-10B

Try LFM • Documentation • LEAP

Daily delta: +2,379 downloads / +10 likes

Open: https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct

This is a **mixed precision NVFP4 quantization** of zai-org/GLM-4.7-Flash, a 30B-A3B (30B total, 3B active) Mixture-of-Experts model.

Daily delta: +2,272 downloads / +10 likes

Open: https://huggingface.co/GadflyII/GLM-4.7-Flash-NVFP4

Try LFM • Documentation • LEAP • WebGPU demo

Daily delta: +2,223 downloads / +11 likes

Open: https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B

Quantized GGUF versions of the TwinFlow Z-Image Turbo model, for stable-diffusion.cpp.

Daily delta: +1,603 downloads / +14 likes

Open: https://huggingface.co/wbruna/TwinFlow-Z-Image-Turbo-sdcpp-GGUF

> Converted and quantized version of zai-org/GLM-4.7-Flash with help of this fix.

Daily delta: +1,477 downloads / +2 likes

Open: https://huggingface.co/evilfreelancer/GLM-4.7-Flash-GGUF

This model is a fine-tuned version of Qwen/Qwen3-4B-Thinking-2507. It has been trained using TRL.

Daily delta: +1,127 downloads / +0 likes

Open: https://huggingface.co/hf-imo-colab/Qwen3-4B-Thinking-2507-SFT