Library
85,762 resources from 11 public sources. 54 checked by hand; the rest are unreviewed.
Results
-
ByteDance: UI-TARS 7B
ModelsUI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games. Built by ByteDance, it builds upon the UI-TARS framework with reinforcement...
bytedance/ui-tars-1.5-7b -
Qwen3-Omni-30B-A3B-Thinking
Modelsany to any model by Qwen · transformers
Qwen/Qwen3-Omni-30B-A3B-Thinking -
Z.ai: GLM 5V Turbo
ModelsGLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks. It natively handles image, video, and text inputs, excels at long-horizon planning, complex coding,...
z-ai/glm-5v-turbo -
Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with...
moondream/moondream3-preview -
Qwen: Qwen3.8 27B
ModelsQwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...
qwen/qwen3.8-27b -
Perceptron: Perceptron Mk1
ModelsPerceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces detailed visual understanding...
perceptron/perceptron-mk1 -
OmAgent
AI repositories[EMNLP-2024] Build multimodal language agents for fast prototype and production
git clone https://github.com/om-ai-lab/OmAgent.git -
Qwen: Qwen3.5-9B
ModelsQwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture. It uses a unified vision-language design...
qwen/qwen3.5-9b -
Qwen: Qwen3.8 27B (batch)
ModelsQwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...
qwen/qwen3.8-27b:batch -
awesome-japanese-llm
AI repositories日本語LLMまとめ - Overview of Japanese LLMs
git clone https://github.com/llm-jp/awesome-japanese-llm.git -
Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding native visual perception and advanced visual...
inclusionai/ling-3.0-flash-vl -
PixelRAG
AI repositorieshttps://arxiv.org/abs/2606.28344. The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/
git clone https://github.com/StarTrail-org/PixelRAG.git -
pixeltable
AI repositoriesThe backend agents build with - Multimodal database, orchestration, and serving in one file
git clone https://github.com/pixeltable/pixeltable.git -
vectordb-recipes
AI repositoriesResource, examples & tutorials for multimodal AI, RAG and agents using vector search and LLMs
git clone https://github.com/lancedb/vectordb-recipes.git -
Google: Gemma 4 26B A4B
ModelsGemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
google/gemma-4-26b-a4b-it -
Hunyuan-MT-7B
Modelstranslation model by tencent · transformers
tencent/Hunyuan-MT-7B -
deeplake
AI repositoriesDeeplake is AI Data Runtime for Agents. It provides serverless postgres with a multimodal datalake, enabling scalable retrieval and training.
git clone https://github.com/activeloopai/deeplake.git -
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
google/gemma-4-26b-a4b-it:free -
Qwen: Qwen3.5-27B
ModelsThe Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance. Its overall capabilities are comparable to those of...
qwen/qwen3.5-27b -
Moondream: Moondream 3.1
ModelsMoondream 3.1 is a vision-language model from Moondream, the successor to Moondream 3 Preview, with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and...
moondream/moondream3.1-9b-a2b -
Hunyuan-1.8B-Instruct
Modelstext generation model by tencent · transformers
tencent/Hunyuan-1.8B-Instruct -
HunyuanDiT
Modelsmodel by Tencent-Hunyuan · hunyuan-dit
Tencent-Hunyuan/HunyuanDiT -
ReForm-Eval
Datasets & benchmarksAn benchmark for evaluating the capabilities of large vision-language models (LVLMs)
git clone https://github.com/FudanDISC/ReForm-Eval.git -
LLaVA-OneVision-2
AI repositoriesFully Open Framework for Democratized Multimodal Training
git clone https://github.com/EvolvingLMMs-Lab/LLaVA-OneVision-2.git
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.