ActIO-UI-7B-RLVR is a 7-billion-parameter visual language model released by Uniphore, specifically designed for computer interface automation tasks. It is based on Qwen2.5-VL-7B-Instruct and optimized through supervised fine-tuning and reinforcement learning with verifiable rewards. It performs excellently in tasks such as GUI navigation, element positioning, and interaction planning, and has achieved the leading level among open-source 7B models in the WARC-Bench benchmark test.

MultimodalTransformersEnglish

This is an open-source large language model with 20 billion parameters focusing on the field of network security. It is based on the GPT-OSS architecture and fine-tuned by integrating multiple network security instruction datasets. The model provides various quantization versions, which are convenient for deployment in resource-constrained environments and can be used to generate network security-related texts, codes, and analysis reports.

Natural Language ProcessingTransformersEnglish

OpenThinker-Agent-v1-SFT is an agent model obtained through supervised fine-tuning (SFT) based on Qwen/Qwen3-8B. It is the first-stage model of the complete training process (SFT + RL) of OpenThinker-Agent-v1, specifically optimized for agent tasks (such as terminal operations and code repair).

Natural Language ProcessingTransformersOther

OpenThinker - Agent - v1 is an open - source intelligent agent model that is subsequently trained based on Qwen3 - 8B. It is designed for terminal operations and software engineering tasks. It is first trained on a high - quality supervised fine - tuning dataset and then further optimized through reinforcement learning. It performs excellently in agent benchmark tests such as Terminal - Bench 2.0 and SWE - Bench.

Natural Language ProcessingTransformersOther

VoxCPM is an innovative tokenizer-free end-to-end text-to-speech (TTS) system that overcomes the limitations of discrete tokenization by modeling speech in a continuous space. It has two core capabilities: context-aware speech generation and realistic zero-shot voice cloning. It can automatically adjust the prosody and style according to the text content and clone the speaker's timbre, accent, and emotion with just a short reference audio.

Audio ProcessingSafetensorsMultiple Languages

LongCat Image Editing is the image editing version of Longcat-Image. It is an open-source image editing model that supports both Chinese and English. It has reached the advanced level of open-source models in terms of instruction following, image quality, and visual consistency, and has excellent precise editing and consistency preservation capabilities.

MultimodalTransformersMultiple Languages

Rnj-1-instruct is an 8-billion-parameter open-weight dense model trained from scratch by Essential AI. It is optimized for code, STEM fields, agent capabilities, and tool invocation. It performs excellently in multiple programming languages and also demonstrates strong capabilities in mathematics and science, comparable to the most advanced open-weight models.

Natural Language ProcessingTransformers

INTELLECT-3 is a Mixture of Experts (MoE) model with 106 billion parameters, trained through large-scale reinforcement learning. It demonstrates excellent performance in mathematics, coding, and reasoning benchmark tests. The model, training framework, and environment are all open-sourced under a permissive license.

Natural Language ProcessingTransformersEnglish

Zenyx_114M-Tiny-Edu-Instruct is an experimental small instruction fine-tuned language model with approximately 114 million parameters. It is built on the TinyEdu-50M base model, pre-trained on the FineWeb-Edu dataset, and fine-tuned on a mixed dataset of OpenHermes-2.5 and CodeFeedback-Filtered. This model aims to explore the limits of instruction fine-tuning under a minimal architecture and verify that the loss converges to approximately 1.04.

Natural Language ProcessingPytorchEnglish

Gemma 3n E4B IT is a lightweight multimodal open model launched by Google, built based on the same research as the Gemini model. This model supports text, audio, and visual inputs and is suitable for various tasks. It uses the MatFormer architecture to achieve efficient parameter utilization.

MultimodalTransformers

TweeTaal-nl-en-0.6B is a language model fine-tuned specifically for Dutch-English bidirectional translation tasks, developed based on the Qwen3-0.6B architecture. This model can still provide accurate and fluent translation services in resource-constrained environments, supporting mutual translation between Dutch and English.

Natural Language ProcessingTransformersOther

Gemma 3 is a lightweight open-source multimodal model launched by Google. This version is an instruction-tuned quantization-aware training model with 12B parameters, which has been converted to the MXFP4 format of the MLX framework. It supports text and image input and generates text output, with a 128K context window and support for over 140 languages.

MultimodalMlx

This is an open-source LoRA model specifically designed for Qwen-Image, focusing on simulating the realistic look and feel of modern iPhone photography. The model is trained on more than 5000 real iPhone-style photos and can generate clear, natural images suitable for social media sharing.

Computer VisionDiffusers

Olmo-3-Think is a fully open-source language model series launched by the Allen Institute for Artificial Intelligence, available in two scales: 7B and 32B. This model is specially trained to demonstrate clear reasoning chains, support transparent step-by-step reasoning and inspectable intermediate thinking traces, and performs excellently in reasoning, mathematics, and code tasks.

Natural Language ProcessingTransformersEnglish

Atom-Olmo3-7B is a language model fine-tuned based on Olmo-3-7B-Instruct, specifically designed for collaborative problem-solving and creative exploration. This model can provide thoughtful and structured analysis when dealing with complex problems, while maintaining an engaging conversation style. It has an Apache 2.0 open-source license.

Natural Language ProcessingTransformersEnglish

SAM 3 is the third-generation promptable segmentation foundation model launched by Meta, which uniformly supports image and video segmentation tasks. Compared with its predecessor SAM 2, it introduces the ability of open vocabulary concept segmentation, can handle a large number of text prompts, and achieves 75-80% of human performance in the SA-CO benchmark test.

Computer VisionTransformersEnglish

The first truly open-source and unrestricted anime video generation model, based on the Wan2.2-5B architecture, can run with only 6GB of video memory and can generate amazing anime animation content.

MultimodalDiffusers

OpenMMReasoner is a completely transparent two-stage multimodal reasoning solution, trained using supervised fine-tuning (SFT) and reinforcement learning (RL). A cold-start dataset of 874K samples was constructed in the SFT stage, and the RL stage used 74K samples to further enhance its capabilities, showing excellent performance in multimodal reasoning benchmark tests.

MultimodalTransformers

OpenMMReasoner is a fully transparent two-stage multimodal reasoning solution that covers supervised fine-tuning (SFT) and reinforcement learning (RL). By carefully constructing high-quality datasets, this solution surpasses powerful baseline models in multiple multimodal reasoning benchmark tests, laying a solid empirical foundation for future large-scale multimodal reasoning research.

MultimodalTransformers

The MiMo Embodied Model (MiMo-Embodied) is a powerful cross-embodied vision-language model that has demonstrated excellent performance in both autonomous driving and embodied AI tasks. It is the first open-source vision-language model that combines these two key domains, significantly enhancing the understanding and reasoning abilities in dynamic physical environments.

MultimodalTransformers