# AI model trends — GitNova https://gitnova.dev/en/ai Source: Hugging Face trending top 50. Source trending rank. Likes are cumulative; downloads cover the last month for the linked version. Quantized copies are grouped; fine-tunes are separate. First seen does not mean a new release. Observed: 2026-10-13T20:52:49+00:00 ## #1 google/embeddinggemma-2 Open multimodal embedding model from Google DeepMind that maps text, code, images, video, and audio into a unified 768-dimensional vector space. Runs on mobile devices and laptops for search, RAG, classification, and clustering. - Model card: https://huggingface.co/google/embeddinggemma-2 - Task: feature-extraction - Likes: 2089 - Downloads in the last month: 124930 - Created: 2026-09-14T14:53:42.000Z - First observed by GitNova: 1791917841 ## #2 jialinyyzz/humanizer A local 12B model that rewrites AI-written drafts (emails, essays, reports, forum posts; English and Chinese) so they read as if a person wrote them, preserving numbers, dates, names and quotes. - Model card: https://huggingface.co/jialinyyzz/humanizer - Task: text-generation - Likes: 1486 - Downloads in the last month: 54039 - Created: 2026-09-11T05:00:16.000Z - First observed by GitNova: 1791917841 ## #3 Qwen/Qwen-Image-2.1-Turbo An accelerated Qwen-Image-2.1 checkpoint for text-to-image generation and image editing in just 8 denoising steps. It loads via QwenImage21Pipeline in Diffusers without manual scheduler configuration. - Model card: https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo - Task: text-to-image - Likes: 701 - Downloads in the last month: 9665 - Created: 2026-10-09T04:50:01.000Z - First observed by GitNova: 1791917841 ## #4 Qwen/Qwen-Image-2.1 Quantized builds of the Qwen-Image-2.1 text-to-image model for local generation in ComfyUI, including uncensored variants without a built-in content filter. The repo ships GGUF, Safetensors and MLX files plus companion text encoder and VAE. - Model card: https://huggingface.co/abenzerps/Qwen-Image-2.1-Uncensored-Quantized - Task: text-to-image - Likes: 4195 - Downloads in the last month: 2247929 - Created: 2026-09-20T15:51:13.000Z - First observed by GitNova: 1791917841 ## #5 Lightricks/LTX-2.5 Image to video - Model card: https://huggingface.co/Lightricks/LTX-2.5 - Task: image-to-video - Likes: 7615 - Downloads in the last month: 1724548 - Created: 2026-07-23T07:55:24.000Z - First observed by GitNova: 1791917841 ## #6 openbmb/MiniCPM-V-4.7-35B-A3B MiniCPM-V 4.7 in the 35B-A3B variant is the first open-source MoE model in the series, handling single-image, multi-image and video understanding and designed for on-device inference. With Lookahead Expert Scheduling, layer-wise quantization and speculative sampling it runs on iPhone at 13+ tokens/s using just 3 GB of memory. - Model card: https://huggingface.co/openbmb/MiniCPM-V-4.7-35B-A3B - Task: image-text-to-text - Likes: 453 - Downloads in the last month: 630 - Created: 2026-10-07T07:07:30.000Z - First observed by GitNova: 1791917841 ## #7 ConwayResearch/Underdog-Saluki-27B-1.0 A 7.89 GB GGUF quantization of Qwen3.8-27B for stock llama.cpp that keeps tool calling intact and runs on modest hardware. An optional vision add-on enables image input. - Model card: https://huggingface.co/ConwayResearch/Underdog-Saluki-27B-1.0 - Task: text-generation - Likes: 417 - Downloads in the last month: 269829 - Created: 2026-10-08T01:47:26.000Z - First observed by GitNova: 1791917841 ## #8 Cloudflare/clef A 27B multimodal model that takes a state (text, JSON, images, video) plus a schema of typed questions and returns probabilities for every allowed option in a single forward pass, with no free-form text generation. - Model card: https://huggingface.co/Cloudflare/clef - Task: image-text-to-text - Likes: 2139 - Downloads in the last month: 17173 - Created: 2026-09-30T21:15:35.000Z - First observed by GitNova: 1791917841 ## #9 LiquidAI/d1-3B A 3B-parameter multimodal decision model built on LFM2.5-VL-3B. It takes a state (text, JSON, images) and a set of questions, returning calibrated typed answers in one forward pass with zero output tokens. - Model card: https://huggingface.co/LiquidAI/d1-3B - Task: image-text-to-text - Likes: 375 - Downloads in the last month: 13190 - Created: 2026-10-05T09:55:28.000Z - First observed by GitNova: 1791917841 ## #10 Qwen/Qwen3-VL-8B-Instruct GGUF builds of the Qwen3-VL-8B-Instruct text encoder with its refusal direction projected out, for running Qwen-Image 2.1 Turbo locally via stable-diffusion.cpp. Includes quantized encoder variants and mmproj. - Model card: https://huggingface.co/AtomicChat/Qwen-Image-2.1-Turbo-Uncensored-GGUF - Task: text-to-image - Likes: 404 - Downloads in the last month: 208575 - Created: 2026-10-09T16:30:48.000Z - First observed by GitNova: 1791917841 ## #11 Qwen/Qwen3.8-27B Open vision-language model Qwen3.8-27B (27B params, up to 1M token context) for coding, agentic tasks, and image/video understanding. Weights in Hugging Face Transformers format, compatible with vLLM, SGLang, and TokenSpeed. - Model card: https://huggingface.co/Qwen/Qwen3.8-27B - Task: image-text-to-text - Likes: 17635 - Downloads in the last month: 7119569 - Created: 2026-08-05T08:22:59.000Z - First observed by GitNova: 1791917841 ## #12 nerkyor/Qwen3.8-27B-Coder390-EfficientThink-Opus5.5-GPT6Astra-Grok4.7-DSV4Pro-K3-SFT-RLOO-MTP-DFlash2 A post-trained multimodal model based on Qwen3.8-27B that fixes the habit of endlessly re-deriving an answer until hitting the context cap. The repo ships weights in many precisions (BF16, FP8, NVFP4, INT8/INT4, GGUF) with SGLang, vLLM, llama.cpp and NInfer support. - Model card: https://huggingface.co/nerkyor/Qwen3.8-27B-Coder390-EfficientThink-Opus5.5-GPT6Astra-Grok4.7-DSV4Pro-K3-SFT-RLOO-MTP-DFlash2 - Task: image-text-to-text - Likes: 304 - Downloads in the last month: 83887 - Created: 2026-10-04T10:44:11.000Z - First observed by GitNova: 1791917841 ## #13 canberkkkkkk/ema-lightning Turkish text-to-speech model: a 5.6M-parameter DiT acoustic model plus a 3M-parameter vocoder, 8.6M total (~34 MB), Apache 2.0. Runs fully offline on GPU or CPU, supports streaming generation and batched calls. - Model card: https://huggingface.co/canberkkkkkk/ema-lightning - Task: text-to-speech - Likes: 403 - Downloads in the last month: 22641 - Created: 2026-10-01T14:18:49.000Z - First observed by GitNova: 1791917841 ## #14 autotrust/JEV-35B Open-weights 35B-parameter decision model (about 3B active per token) built on Qwen3.5-35B-A3B, designed to pick one calibrated answer from up to 256 options in a single pass for ranking, classification and search re-ranking. - Model card: https://huggingface.co/autotrust/JEV-35B - Task: text-classification - Likes: 412 - Downloads in the last month: 17426 - Created: 2026-10-11T01:22:03.000Z - First observed by GitNova: 1791917841 ## #15 baidu/Unlimited-OCR Baidu's OCR model for text recognition on images and multi-page PDFs, handling long documents in a single pass. Runs via Transformers, vLLM, or SGLang. - Model card: https://huggingface.co/baidu/Unlimited-OCR - Task: image-text-to-text - Likes: 4609 - Downloads in the last month: 1175446 - Created: 2026-06-19T09:40:33.000Z - First observed by GitNova: 1791917841 ## #16 convaiinnovations/laya Multilingual, non-autoregressive decision model that takes state (text, email, ticket, or JSON) and typed questions and returns answers with calibrated probabilities in one forward pass. Trained to report honest probabilities, so it doesn't generate text and can't hallucinate. - Model card: https://huggingface.co/convaiinnovations/laya - Task: text-classification - Likes: 5572 - Downloads in the last month: 46160 - Created: 2026-09-18T05:05:55.000Z - First observed by GitNova: 1791917841 ## #17 speridlabs/iris-3b Iris-3B is a 3B-parameter pixel-space image generator that outputs pixels directly without a VAE or latent space, plus fine-tuned variants for depth estimation and image restoration/upscaling. A Turbo version generates 1024x1024 images in about 0.7s on an H100. - Model card: https://huggingface.co/speridlabs/iris-3b - Task: text-to-image - Likes: 231 - Downloads in the last month: 889 - Created: 2026-10-05T14:51:15.000Z - First observed by GitNova: 1791917841 ## #18 Cactus-Compute/whistle Whistle is a compact speech recognition model for mobiles, wearables, robots, and microcontrollers. It runs on CPU without GPU or dependencies, supporting transcription, word timestamps, and speech embeddings. - Model card: https://huggingface.co/Cactus-Compute/whistle - Task: automatic-speech-recognition - Likes: 380 - Downloads in the last month: 21005 - Created: 2026-09-30T17:13:26.000Z - First observed by GitNova: 1791917841 ## #19 openbmb/MiniCPM-V-4.7-1B MiniCPM-V 4.7 is a 1B multimodal model for single-image, multi-image and video understanding, optimized for on-device and local inference. - Model card: https://huggingface.co/openbmb/MiniCPM-V-4.7-1B - Task: image-text-to-text - Likes: 280 - Downloads in the last month: 263 - Created: 2026-10-06T17:40:13.000Z - First observed by GitNova: 1791917841 ## #21 XingChen-AGI/Xing4.0-29B-A4B Text and chat - Model card: https://huggingface.co/Venastine-Research/Xing4.0-29B-A4B-GGUF - Task: text-generation - Likes: 709 - Downloads in the last month: 58000 - Created: 2026-09-17T18:37:31.000Z - First observed by GitNova: 1791917841 ## #23 facebook/sam3 Other tasks - Model card: https://huggingface.co/facebook/sam3 - Task: mask-generation - Likes: 4246 - Downloads in the last month: 2037775 - Created: 2025-11-07T05:17:48.000Z - First observed by GitNova: 1791917841 ## #24 DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NM-DAU Vision and chat - Model card: https://huggingface.co/DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF - Task: image-text-to-text - Likes: 1733 - Downloads in the last month: 1940062 - Created: 2026-09-01T07:01:09.000Z - First observed by GitNova: 1791917841 ## #25 Aleph-Alpha/Kolibri-1 Text and chat - Model card: https://huggingface.co/Aleph-Alpha/Kolibri-1 - Task: text-generation - Likes: 926 - Downloads in the last month: 15335 - Created: 2026-10-02T05:54:02.000Z - First observed by GitNova: 1791917841 ## #26 Qwen/Qwen3.8-Flash-Next Vision and chat - Model card: https://huggingface.co/ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF - Task: image-text-to-text - Likes: 853 - Downloads in the last month: 4675423 - Created: 2026-09-07T09:41:37.000Z - First observed by GitNova: 1791917841 ## #27 Cloudflare/clef-flash Vision and chat - Model card: https://huggingface.co/Cloudflare/clef-flash - Task: image-text-to-text - Likes: 803 - Downloads in the last month: 24809 - Created: 2026-09-30T21:15:36.000Z - First observed by GitNova: 1791917841 ## #29 pablodawson/MiniMax-H3-360-Orbit-LoRA Other tasks - Model card: https://huggingface.co/pablodawson/MiniMax-H3-360-Orbit-LoRA - Task: image-text-to-video - Likes: 413 - Downloads in the last month: 15540 - Created: 2026-09-27T04:05:33.000Z - First observed by GitNova: 1791917841 ## #31 deepseek-ai/DeepSeek-V4.1-Flash Vision and chat - Model card: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash - Task: image-text-to-text - Likes: 4467 - Downloads in the last month: 1272612 - Created: 2026-09-10T02:17:58.000Z - First observed by GitNova: 1791917841 ## #37 vllm-sr/d3 Other tasks - Model card: https://huggingface.co/vllm-sr/d3 - Task: zero-shot-classification - Likes: 135 - Downloads in the last month: 1156 - Created: 2026-10-10T09:38:30.000Z - First observed by GitNova: 1791917841 ## #38 pyannote/speaker-diarization-community-1 Speech recognition - Model card: https://huggingface.co/pyannote/speaker-diarization-community-1 - Task: automatic-speech-recognition - Likes: 2848 - Downloads in the last month: 5444861 - Created: 2025-04-15T21:31:35.000Z - First observed by GitNova: 1791917841 ## #39 Comfy-Org/Qwen-Image-2.1 Other tasks - Model card: https://huggingface.co/Comfy-Org/Qwen-Image-2.1 - Task: other - Likes: 1102 - Downloads in the last month: 9400716 - Created: 2026-09-15T19:41:09.000Z - First observed by GitNova: 1791917841 ## #40 perplexity-ai/pplx-decider-v1.1-27b Other tasks - Model card: https://huggingface.co/perplexity-ai/pplx-decider-v1.1-27b - Task: text-classification - Likes: 141 - Downloads in the last month: 2853 - Created: 2026-10-05T12:32:27.000Z - First observed by GitNova: 1791917841 ## #41 zai-org/GLM-5.3-Flash Vision and chat - Model card: https://huggingface.co/autotrust/GLM5.3-Flash-E192-DGX-Spark - Task: image-text-to-text - Likes: 175 - Downloads in the last month: 4537 - Created: 2026-10-09T23:02:37.000Z - First observed by GitNova: 1791917841 ## #42 LiquidAI/LFM2.5-24B-A2B Text and chat - Model card: https://huggingface.co/LiquidAI/LFM2.5-24B-A2B - Task: text-generation - Likes: 121 - Downloads in the last month: 2565 - Created: 2026-10-12T05:35:46.000Z - First observed by GitNova: 1791921146 ## #43 JetBrains/Mellum2.1-12B-A2.5B-Thinking Text and chat - Model card: https://huggingface.co/JetBrains/Mellum2.1-12B-A2.5B-Thinking - Task: text-generation - Likes: 120 - Downloads in the last month: 1471 - Created: 2026-09-20T10:52:55.000Z - First observed by GitNova: 1791917841 ## #45 LiquidAI/d1-omni-600M Vision and chat - Model card: https://huggingface.co/LiquidAI/d1-omni-600M - Task: image-text-to-text - Likes: 118 - Downloads in the last month: 11514 - Created: 2026-10-05T17:18:48.000Z - First observed by GitNova: 1791917841 ## #46 ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF Vision and chat - Model card: https://huggingface.co/SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF - Task: image-text-to-text - Likes: 240 - Downloads in the last month: 1168151 - Created: 2026-10-02T03:42:33.000Z - First observed by GitNova: 1791917841 ## #47 Cloudflare/clef-omni Vision and chat - Model card: https://huggingface.co/Cloudflare/clef-omni - Task: image-text-to-text - Likes: 116 - Downloads in the last month: 1195 - Created: 2026-10-09T04:10:22.000Z - First observed by GitNova: 1791917841 ## #48 Alissonerdx/BFS-Best-Face-Swap Image editing - Model card: https://huggingface.co/Alissonerdx/BFS-Best-Face-Swap - Task: image-to-image - Likes: 1397 - Downloads in the last month: 261734 - Created: 2025-11-07T09:50:28.000Z - First observed by GitNova: 1791917841 ## #49 MiniMaxAI/MiniMax-H3 Other tasks - Model card: https://huggingface.co/MiniMaxAI/MiniMax-H3 - Task: image-text-to-video - Likes: 6113 - Downloads in the last month: 3697694 - Created: 2026-07-28T10:45:18.000Z - First observed by GitNova: 1791917841 ## #50 akatz-ai/MiniMax-H3-Character-Swap-LoRA Other tasks - Model card: https://huggingface.co/akatz-ai/MiniMax-H3-Character-Swap-LoRA - Task: video-to-video - Likes: 445 - Downloads in the last month: 48236 - Created: 2026-09-25T08:14:15.000Z - First observed by GitNova: 1791917841