Latest AI Resources

OpenReasoning-Nemotron is a series of large-scale language models open-sourced by NVIDIA to support processing of reasoning tasks in math, science and code. The models are distilled based on the DeepSeek R1 0528 model with parameter scales of 1.5B...

Seed-X is a multilingual translation model launched by the Seed team of ByteDance, with 7 billion parameters, supporting two-way translation in 28 languages. The model combines multilingual data pre-training, command fine-tuning and reinforcement learning techniques to efficiently handle complex language patterns and make translation quality better...

JoyAgent-JDGenie is Jingdong open source lightweight general multi-intelligence system , no need for secondary development can be used directly.JoyAgent-JDGenie can handle complex tasks , such as generating reports , analyzing data , etc. , supports a variety of delivery formats , such as web pages , PPT ...

TRAE SOLO is an AI automated development assistant introduced by TRAE, an AI programming assistant launched by ByteDance, to simplify the software development process with AI technology.TRAE SOLO understands the user's needs, supports text descriptions, voice commands, and file uploads to input the requirements, and automatically plans...

Fogsight is an innovative AI animation generation agent that transforms abstract concepts into vivid animations based on large-scale language modeling (LLM). Users input topics and Fogsight generates complete animations with bilingual narration and cinematic visuals.

Goedel-Prover-V2 is an open-source theorem proving model jointly released by leading organizations such as Princeton University, Tsinghua University, and NVIDIA. The model is based on innovative techniques such as hierarchical data synthesis, verifier-guided self-correction, and model averaging to significantly improve the performance of automated formal proofs...

BytePlus is an enterprise-level intelligent service platform launched by BytePlus to provide diversified services overseas. The platform covers powerful functions such as content distribution and acceleration (CDN), personalized recommendation, augmented reality, data processing and analysis, real-time audio and video communication, artificial intelligence and machine learning.

Flying Book Miaohu is an enterprise-level AI native system building platform launched by Flying Book. The platform quickly transforms enterprise business requirements into practical applications through a multi-agent architecture, supporting the whole process from requirements analysis to functional design, application development and problem repair. Users use a dialog to easily build lightweight...

MirageLSD is the world's first real-time streaming diffusion AI video model from the Decart AI team, enabling unlimited real-time video generation with latency as low as 40 milliseconds and smooth output at 24 frames/second.

Kimi Playground is an AI tool-calling experience platform for developers from Dark Side of the Moon.Kimi Playground enables AI to call a variety of tools (e.g., weather lookups, hotel bookings, data analytics, etc.) to accomplish complex tasks, not just...

ChatGPT Agent is a general-purpose AI Agent from OpenAI that combines multiple capabilities to autonomously accomplish complex tasks. Users only need to describe their needs in natural language, and the Agent can automatically select the appropriate tools, such as browsing the web, extracting information, running code...

GLM-Experimental is an experimental large language model launched by Wisdom Spectrum AI, which has been launched on the Z.ai platform. The model has the ability to automatically generate PPTs. After users input topics or key points, the model can quickly generate clearly structured and standardized presentations, which will...

PromptForge is a powerful AI cue engineering workbench that helps users efficiently create, test, and optimize AI cues.PromptForge provides optimization suggestions through intelligent generation features that support multi-model comparisons and dynamic parameter tuning, giving users the flexibility to compare no...

VikingDB is a high-performance cloud-native vector database launched by Volcano Engine, designed for processing massive high-dimensional vector data.VikingDB has a variety of data writing methods, including real-time synchronization, asynchronous writing, etc., to meet the data processing needs of different scenarios.VikingD...

RoboBrain 2.0 is an open source advanced embodied brain model that provides powerful perception, reasoning, and planning capabilities for robots.RoboBrain 2.0 consists of two versions, 7B and 32B, and utilizes a heterogeneous architecture that fuses visual encoders and language models to support...

RoboOS 2.0 is an open-source framework for cross ontology brain collaboration, promoting the transformation of robots from single intelligence to collaborative group intelligence. The framework realizes efficient division of labor with a "big brain" architecture, where the cloud brain is responsible for complex decision-making and collaboration, and the small brain module focuses on executing specific skills.

EXAONE 4.0 is a hybrid reasoning grand model from LG AI Research in Korea, blending general-purpose natural language processing and advanced reasoning capabilities. The model supports Korean, English, and Spanish, and is divided into a 32B professional version and a 1.2B end-side version. The professional version is suitable for legal, accounting...

Tizzy.ai is an AI intelligent search application launched by Baidu.Tizzy.ai is based on Baidu's big model technology, with powerful intelligent search functions, can quickly answer questions, deep thinking and assist in decision-making.Tizzy.ai has a simple interface, no ads and pop-ups, and the bottom of the guide...

Voxtral is a state-of-the-art open source speech model from Mistral AI that supports the advancement of natural human-computer interaction through powerful speech transcription and comprehension capabilities. The model is available in two versions, 24B for mass production and 3B for local deployment.Voxtral supports...

k2 is a MoE architecture base model with superb code and Agent capabilities from Moonshot AI, with 1T total parameters and 32B activation parameters. in benchmark performance tests in the main categories of General Knowledge Reasoning, Programming, Mathematics, and Agent, the k2 model...

Grok 4 is the latest big AI model from xAI, and Grok 4 delivers a 10x improvement in reasoning power over its predecessor. The model's superior reasoning ability enables it to score near perfect on difficult exams such as the SAT and GRE, and outperforms other cutting-edge models in a number of benchmark tests...

GenFlow Super Hitchhiker is a general-purpose AI Agent launched by Baidu Literature Library, which allows users to autonomously disassemble tasks, call up Baidu Literature Library's 1.4 billion document libraries and online resources, and generate PPTs, reports, charts, posters, and other full-modal content in an extremely fast manner by simply typing in the natural language commands.

TRAE Agent is ByteDance's open source intelligent software engineering assistant, built on advanced Large Language Model (LLM).TRAE Agent efficiently handles complex software development tasks, including codebase exploration, issue reproduction, fix formulation, and high-quality code generation, etc....

Higress MCP is an innovative platform launched by Invest Today that supports the rapid transformation of traditional financial data APIs into modern MCP services.Higress MCP enables the transformation of REST APIs to MCP Server based on a simple configuration without the need to program...

CodeBuddy IDE is an AI-driven full-stack development assistant launched by Tencent, based on intelligent means to improve development efficiency.CodeBuddy IDE has a powerful AI code completion function, which supports real-time prediction and automatic completion of code fragments to reduce repetitive labor.C...

Pet TA is an AIGC pet interactive product launched by Jingdong, which can provide a fun and cozy online interactive platform for pet lovers. It supports users to choose a variety of cute clothes and accessories for their pets, personalized dress up, and can create a digital image of their pets for rich interaction with them. The platform provides...

Chat Healing Small Universe is an AI mental companion product launched by Jingdong Health. Launched on the basis of the big medical model of Jingdao Qianxin, Chat Healing Small Universe can simulate the human communication process and empathy, providing users with 7×24 hours of mental health support. Users interact with the AI character Little Fish based on text, emoji, or voice to share emo...

Shortcut is an innovative AI-driven Excel assistance tool. Based on intelligent algorithms, it can quickly complete a variety of complex Excel tasks, including data entry, editing, calculating, and organizing.Shortcut can handle basic operations and support responding to advanced gold...

StarFlow Agent is a one-stop creative design agent designed for Chinese users by Liblib.StarFlow Agent continues Lovart's full-stack intelligent design capabilities, fully adapted to Chinese semantics, oriental aesthetics and local scenarios. Users only need to enter a sentence, Agen...

Step-Audio-AQAA is an end-to-end large-scale audio language model for Audio Query-Audio Answer (AQAA) tasks from the StepFun team. It can directly process audio input to generate natural and accurate speech responses without relying on traditional automatic speech recognition (A...

MuseSteamer is a large model for multimodal video generation launched by Baidu. The model can quickly generate high-quality dynamic video content based on text descriptions or images provided by the user, supporting a variety of clarity and functionality versions to meet the needs of different scenarios of creation.

Painting is an AI video generation platform launched by Baidu, based on AI technology to help users easily create personalized videos. Painting intuitive interface, powerful tools, with inspiration recommendation function, can provide creators with creative inspiration, support a key to the same operation, can quickly generate similar videos, simplify the creative process.

Antelope is an intelligent design tool launched by Jingdong, providing efficient and convenient design solutions for e-commerce merchants and individuals. Through intelligent keying, intelligent layout, intelligent color matching and other functions, it helps users to quickly generate high-quality design works to meet the main picture of the product, advertising Banner, store page and other e-commerce store...

Tencent Advertising Myth is a one-stop AI advertising creative platform launched by Tencent Advertising, based on Tencent's hybrid big model, providing advertisers with efficient and intelligent creative generation and optimization services. It supports a variety of creative tools such as text-to-map, map-to-map, text-to-video, etc. Users only need to enter a simple natural language description or upload material...

Tucoli is an AI image processing tool that focuses on providing efficient and convenient photo and video editing services. Based on advanced artificial intelligence technology, it can quickly realize a variety of functions such as one-click keying, video processing, photo repair and beautification, AI painting and background replacement.

Circle Traveler is AI travel planning app that helps users arrange travel itineraries efficiently and conveniently. Users can quickly enter information such as destination and time to automatically generate a reasonable and personalized itinerary. It supports one-click import of links, text or images from social platforms to quickly generate the same itinerary.

ML-Master is an AI expert intelligences launched by Agents team of Artificial Intelligence School of Shanghai Jiaotong University. It performed well in OpenAI's authoritative benchmark test MLE-bench, topping the list with an average medal rate of 29.3%, surpassing Microsoft's RD-Agent...

Paperpal is an AI aid focused on academic writing, helping researchers, students and professionals to improve the quality and efficiency of their writing. With powerful language touch-up features, it provides grammar checking, spelling correction, and word choice suggestions to ensure accurate and fluent text.

GLM-4.1V-Thinking is an open source visual language model introduced by Smart Spectrum AI, designed for complex cognitive tasks.GLM-4.1V-Thinking supports multimodal inputs, covering images, videos and documents. Based on the GLM-4V architecture, the model introduces a chain of thought...

ThinkSound is the first CoT (Chain Thinking) audio generation model introduced by Ali Tongyi's speech team. The model can generate accurately matched sound effects for video images, based on the introduction of CoT reasoning, to solve the problem of traditional technology is difficult to capture the dynamic details of the screen and spatial relationships.

Mapify is an AI mind map generator tool launched by XMind team. It can quickly convert text, PDF, web pages, video, audio and other formats into structured mind maps, helping users efficiently extract and organize key information.

CodeArts Doer is an AI intelligent development assistant launched by Huawei Cloud to improve development efficiency and software quality through AI technology. CodeArts Doer is based on 6 Agent Intelligences (Team, Project, Product, Developer, Test, and Committer Assistant), covering from the need...

Qwen-TTS is an advanced speech synthesis model introduced by Ali Tongyi. The model can efficiently convert text into natural and smooth speech, supporting multiple languages and dialects, such as Mandarin, English, Beijing dialect, etc., to meet the needs of different regions and scenes. Relying on massive corpus training, the model's speech output is of high quality, rhyming...

MultiAgentPPT is an open source multi-intelligent AI presentation generation system. Users only need to enter the subject , the system is based on multi-intelligent collaboration , automatically complete the outline generation , subject splitting , parallel research and content summarization and other steps to quickly generate high-quality PPT....

Ovis-U1 is a multimodal unified model introduced by the Ovis team of Alibaba Group with a parameter scale of 3 billion. The model is equipped with three core capabilities: multimodal understanding, text-to-image generation, and image editing. With advanced architectural design and collaborative and unified training methods, the model supports the realization of high-fidelity image...

Kangaroo Staff is a merchant-oriented AI intelligent decision-making application launched by Meituan to help merchants solve problems in store opening and operation. Based on Meituan's massive catering data and more than 10 years of online operation experience, through conversational interaction, it provides merchants with precise information on track selection, store opening location, dish development, store operation and other scenarios, such as...

AI Jitterbug is an intelligent deep thinking and search application launched by Jitterbug to provide users with a more efficient and intelligent content acquisition experience. Based on Jitterbug's powerful content ecosystem and AI technology, it provides users with more comprehensive and detailed answers through connected search and reasoning capabilities.

Qwen VLo is a multimodal unified comprehension and generation model introduced by Tongyi Qianqian team. Qwen VLo can "understand" the world and recreate with high quality based on its understanding, realizing the leap from perception to generation. VLo can accurately understand the image content, and on the basis of this, it can carry out consistent and high-quality generation.

AiBiao is AI chart processing tool that helps users efficiently complete data processing and analysis work through natural language interaction. It supports users to query, clean, merge, calculate and analyze data through conversations without complex formulas or codes, which lowers the threshold of data processing.

Doppl is an AI virtual fitting application launched by Google. After the user uploads a full body photo, the application supports the clothing picture or screenshot "wear" in the digital version of their own body, and can be converted from static pictures to AI-generated video, so that users can more truly feel the effect of clothing on the body.

Xunlei MCP is launched by Xunlei, an automatic download service based on AI technology. Users in the AI application that supports the service, with voice or text input download demand, AI can automatically search for network resources and start downloading. Xunlei MCP supports PC version of Xunlei and NAS Xunlei, breaking the traditional download mode, allowing...

TicNote is an AI voice recorder launched by TicNote, which is a combination of Agentic AI hardware and software products, and is positioned as a "portable AI thinking partner". It adopts card-type design, thin and lightweight, and is equipped with a magnetic protective case, which can be easily carried or attached to the back of the phone.

Wanxing Canopy is the AIGC video creation platform launched by Wanxing Technology, covering the three major creation fields of video, picture and audio generation, which is specially designed for media and cultural industry workers, film and television/post-production workers, art and design workers, advertising and marketing practitioners, etc. to provide one-stop professional creation solutions.

Kapi Bookkeeping is an intelligent AI bookkeeping app launched by Shangtang Technology. The application takes automatic bookkeeping as its core function, automatically recognizes amounts and classifications, and supports voice input, making bookkeeping easy and convenient. Kapi Bookkeeping can intelligently analyze billing data and regularly push personalized consumption summaries and financial advice to help users better...

Gemini CLI is Google's open source AI programming tool based on incorporating the Gemini Big Model into the developer's endpoint to provide developers with powerful AI capabilities. The tool understands code, manipulates files, executes commands, and dynamically troubleshoots problems to help developers efficiently write generation...

AnimaTensor is a quadratic image generation model from the CagliostroLab team and TensorArt, based on an innovative V-Prediction technique that optimizes noise scheduling by predicting the "speed" of the image generation process....

Gemini Robotics On-Device is a vision-language-action model from Google DeepMind that supports running locally in a robot. The model is able to perform tasks offline, performing fine actions based on natural language commands, such as folding clothes and pulling open bags...

11ai is an AI voice assistant launched by ElevenLabs, with voice interaction as the core, through natural and smooth dialogue to enhance the user's work efficiency. 11ai supports more than 5,000 voices, and users can customize the exclusive voice, the assistant is more personalized. With low-latency voice inter...

Inkblade AIPPT is the inkblade launched AI PPT creation tools, based on intelligent technology to simplify the production process of PPT. Users only need to enter the theme or import the document, AI can quickly generate a complete structure, design unity and content-rich PPT. tools with intelligent matching copy, a key to generate charts and diagrams, self...

Mu is a small-parameter language model from Microsoft with only 330 million parameters, designed to run efficiently on NPUs and edge devices. The model is based on an encoder-decoder architecture, and achieves a response rate of over 100 tokens per second through hardware-aware optimization, model quantization, and task-specific fine-tuning.M...

TanRice is an AI food recommendation tool launched by Jitterbug, a subsidiary of ByteDance, which relies on the Beanbag Big Model to provide users with personalized food recommendations, store scouting comparisons, food tips and other services. TanRice can accurately recommend nearby restaurants and dishes based on users' taste preferences and location, support assisted ordering, and provide group-buying and takeaway services...

Confucius3-Math is the first domestic open source reasoning model focused on mathematics education open-sourced by NetEaseYouDao. With 14 billion parameters, optimized for K-12 math education scenarios, it can run efficiently on a single consumer GPU (such as RTX 4090D), with an inference performance of about...

Comate AI IDE is the industry's first multi-modal, multi-intelligence body collaborative AI native IDE launched by Baidu Wenshin Express Code, with powerful multi-modal capabilities, support for the design draft of a key to the code (F2C), image to the code, and the natural language to the code, the front-end development scenarios in the performance of the outstanding...

DeepPiano is an intelligent piano application with a big model as its core launched by Zhiqu Technology. Through advanced artificial intelligence technology, it provides a variety of convenient functions for piano players and learners.DeepPiano can realize intelligent sheet music page turning, automatic recognition of playing progress, no need to manually operate...

MoFlow is an emotion management app that combines psychology and AI technology. It provides a private space for users to express their emotions through writing and chatting, and the AI partner will patiently listen to give feedback.

Kimi-Researcher is a new generation Agent model trained on end-to-end autonomous reinforcement learning (end-to-end agentic RL) technology from Kimi, a subsidiary of Dark Side of the Moon, designed for deep research tasks. It can autonomously regulate...

OmniTalker is a real-time text-driven talking avatar generation framework from Alibaba. The framework can simultaneously handle multiple modal inputs such as text, images, audio and video to generate natural speech responses based on a streaming approach. The core is based on the Thinker-Talker architecture , Thin...

FactSnap is a new generation AI information verification tool that helps users quickly verify the authenticity of web information. By integrating multiple models and search engines, it performs real-time verification of selected text while the user is browsing the web.

SongGeneration is a high-quality song generation project launched by Tencent AI Lab. Based on the LeLM (Large Language Model for Music) framework, it can generate vocals and accompaniment in parallel, realizing the harmony of the two...

DeepSite V2 is an AI web page generation tool based on the DeepSeek R1-0528 model. Without installing or configuring a local environment, users can quickly generate, preview, and iterate website pages with simple text prompts.

Skywork-SWE-32B is an open source 32B scale software engineering (SWE) autonomous code intelligences base model introduced by Kunlun World Wide Web. The model focuses on software engineering tasks, has powerful repository-level code repair capabilities, and can perform in complex scenarios with multi-round interactions and long text processing...

Hailuo Video Agent is a zero-threshold AI video creation Agent launched by MiniMax, open Beta version. Through simple text input or image upload, high-quality creative videos can be generated in one click, covering a variety of application scenarios, such as broad...

AQ is an AI medical health tool launched by Ant Group. Based on advanced AI technology, it provides users with convenient medical health services. The tool is based on AQ for health consultation, access to AI intelligent Q&A services, support for uploading pictures for interpretation of reports, cases, etc. AQ supports personal and family health record management...

Magentic-UI is Microsoft's open source human-computer collaboration AI Agent research tool.Magentic-UI is based on working closely with users to facilitate complex Web tasks such as Web browsing, code execution, and file handling. The tool emphasizes collaborative planning, enabling users to raise...

Flycount JavaAI is an intelligent Java development assistant launched by Flycount Technology. The platform supports natural language input to realize the whole process of intelligent development from requirements analysis to code generation. Developers only need to enter a description of the requirements, Flycount JavaAI can accurately understand and generate a complete engineering code framework, the platform...

Midjourney V1 is the first AI video generation model from Midjourney, which supports transforming static images into vivid and dynamic videos with the help of advanced AI technology. Users just need to upload images or images generated with Midjourney, tap...

xbench is an AI benchmarking tool launched by Sequoia China. Based on the dual-track evaluation system, it evaluates the upper limit of AI system's capability and technology boundary on one hand, and quantifies the utility value of AI system in real scenarios on the other hand. xbench is based on the evergreen mechanism of evaluation, and dynamically updates the test content to ensure that the evaluation...

Love Chart is an intelligent data visualization platform launched by Dysprosium Technology. Based on AI technology, the platform automatically recommends appropriate chart types based on data characteristics and generates beautiful chart styles. The platform supports bar charts, line charts, pie charts and other chart types, and provides rich templates for users to customize.

wcplusPro is a professional WeChat public number data collection and analysis tool. The tool supports the collection of public number history articles and a variety of data (such as title, content, readership, etc.), provides efficient data management functions, and supports handling fast retrieval and sorting of million-level articles. wcplusPro...

Xiaoka Health is an AI health management app launched by Beijing FioTech to help users easily manage calorie intake and consumption. Users just need to take a picture of the food, and the app can quickly recognize the calories and other nutrients of the food and record them automatically without manual input. The app is based on the user's age, gender, weight, height...

Miaojiao is a professional AI video content analysis and generation tool, based on deep learning algorithms, which analyzes the visual, audio and creative structure of the video in an all-round way, revealing the laws behind the explosive videos. Miaojiao can analyze composition, color, and lens language, assess creative uniqueness and emotional resonance, and provide trend insights and optimization suggestions...

ChartAI is an efficient data visualization tool that automatically analyzes data based on AI technology and generates intuitive charts to help users quickly understand complex information.ChartAI supports the import of a variety of data formats (e.g., Excel, CSV), and supports intelligent recommendation of the most suitable charts ...

Kotae is an intelligent chatbot platform designed for small businesses and developed on ChatGPT technology. Supporting over 80 languages, the platform responds quickly and naturally to customer inquiries, generating accurate answers based on analyzing a user's website content, training documents, FAQs, etc. Kotae has no...

OpenEvidence is a medical knowledge assistant platform based on AI technology to provide accurate clinical support for doctors and healthcare workers. The platform is based on small specialized models and multi-model integration architecture to quickly answer clinical questions, analyze symptoms, recommend treatment plans, and provide the latest medical knowledge more...

Leaning Wonderful Writer is an AI dissertation writing tool that provides an efficient and convenient solution for academic writing. The tool supports one-click generation of dissertation outlines, abstracts and first drafts of the body of the paper, which is applicable to different levels of academic needs such as undergraduate and master's degrees, and covers multi-disciplinary fields such as science and technology, liberal arts and social sciences.

WriteHERE is an open source AI long essay writing framework introduced by a team led by Jürgen Schmidhuber. The framework is based on a heterogeneous recursive planning technique that dynamically decomposes the writing task into three heterogeneous tasks: retrieval, reasoning and writing, based on a directed acyclic graph (DA...

SurveyGO (Volume Ji) is an open source AI paper writing tool from Tsinghua University in conjunction with the Facing Intelligence team, which can help researchers and students to quickly generate literature review sections. Users only need to enter the topic and description of the paper, the tool can quickly generate a clear structure and rich content of the review article, support tens of thousands of words long...

Hailuo 02 is a new AI video generation model from MiniMax, an upgraded version of Hailuo 01. The model performs well in the graph-generated and text-generated video domains, ranking second only to Byte's Seedance 1.0 and surpassing the fast...

Easy Pen AI is an intelligent tool designed for academic writing that helps students, researchers and all kinds of academic writers to quickly generate high-quality first drafts and frameworks for their papers. Based on advanced AI technology, the tool generates in a short time the first draft of a paper containing introduction, literature review, research methodology based on the topic and keywords of the paper inputted by the user...

Mindsera is an intelligent AI journaling tool that helps users improve their mental health and cognitive abilities based on AI technology.Mindsera parses the user's text in depth, tapping into underlying emotions and thought patterns to help users understand themselves better.Mindsera has a built-in smart chat feature...

Color Swirl is a one-stop PPT generation platform that integrates AI intelligent creation, multi-person collaboration and secure sharing. Users input the theme or import text, the platform can generate high-quality presentations with one click, supporting intelligent graphics, text optimization and content layout adjustment. Color Swirl provides a rich library of templates to meet the needs of different scenarios ...

MagicSet is an efficient and convenient online AI design platform that provides users with one-stop design solutions. The platform integrates a rich copyright material library, with 500,000+ genuine pictures, illustrations, art characters and other materials, users can commercialize for free without worrying about copyright issues. The platform provides more than 200,000 design templates, including...

Teacher Help is an AI intelligent tool platform designed for teachers to improve their work efficiency and teaching quality based on AI technology. The platform provides a variety of functions, including lesson plan generation, one-click conversion of courseware to PPT, homework and test question design, student comment generation, and teaching plan writing. The platform supports text translation...

DeckSpeed is an AI presentation creation tool based on conversational interaction, where users express their needs based on natural language and quickly generate personalized slides without relying on traditional templates. The tool supports real-time feedback adjustment, users can modify the color, style and content of the slide at any time to ensure that the presentation is complete...

Slidev is an open source slide production tool , based on Markdown and Vue technology stack development . Tools to support users with simple Markdown syntax to quickly create slides , support for code highlighting , real-time coding demonstration , mathematical formula rendering , chart generation and other functions ...

DeerFlow is a byte jump open source deep research framework designed for efficiently accomplishing complex research tasks. The framework integrates multiple language models (e.g. Qwen) and tools (e.g. search engines , crawlers , Python execution environment ) , can quickly generate high-quality research reports , podcasts ...

Dolphin is a byte jump open source lightweight document parsing large model , with 322M parameters , small size and fast running speed . The model is based on a two-stage parsing approach , based on page-level layout analysis to identify the elements in the document ( such as headings , tables, formulas , etc.) , and then the content of each element parsed ...

PromptPilot is an AI prompt optimization platform launched by Volcano Engine under ByteDance. The platform deeply analyzes the user's intention and transforms vague ideas into professional instructions that can be precisely executed by AI, ensuring that the model outputs high-quality results. Based on interactive guidance, PromptP...

Kimi-Dev is an open source code model from Moonshot AI designed for software engineering with 72B parameters. The model has a powerful BugFixer feature that can automatically locate and fix code errors, and the model provides test code generation...