Search results for "large"

FlagEval (Libra) is a large-scale model evaluation system and open platform initiated by Beijing Zhiyuan Artificial Intelligence Research Institute, aiming to establish scientific, fair, and open evaluation benchmarks and methods. The platform has innovatively constructed a three-dimensional

OpenCompass: An Open Source Large Model Comprehensive Evaluation System and Sinan Open Platform

211

OpenCompass is an open-source large model evaluation system launched by Shanghai Artificial Intelligence Laboratory, providing one-stop evaluation services for large language models, multimodal models, and scientific intelligence models. The platform supports one click distributed evaluation of

AGI Eval: A large model evaluation community and authoritative third-party evaluation platform

208

AGI Eval is a large model evaluation community jointly created by top universities and institutions such as Shanghai Jiao Tong University, Tongji University, East China Normal University, and DataWhale, with the mission of "assisting evaluation and making AI a better partner for humanity". The

CMMLU: Authoritative Chinese Language Model Knowledge Understanding Ability Evaluation Benchmark

192

CMMLU (Chinese Massive Multitask Language Understanding) is a large-scale language understanding benchmark designed specifically for the Chinese language context, covering 67 subject topics from beginner to advanced professional levels, including natural sciences, social sciences, engineering

2026-05-30

MMLU: The International Authoritative Benchmark for Multi Task Language Understanding Ability of

171

MMLU (Massive Multitask Language Understanding) is a large-scale multi task language understanding evaluation dataset jointly released by the University of California, Berkeley and other institutions. It covers 57 disciplinary fields, including humanities, social sciences, natural sciences,

LangGPT - Structured Prompt Word Design Framework: Natural Language Programming for Large Language

162

LangGPT is a structured prompt word design framework inspired by programming languages, aimed at solving the problem of non AI experts having difficulty writing high-quality prompt words. Drawing on the design ideas of object-oriented programming languages, LangGPT proposed a dual layer structure

2026-05-29

HELM: Stanford University led comprehensive evaluation framework for large language models and high

158

HELM (Holistic Evaluation of Language Models) is a comprehensive language model evaluation framework initiated by the Stanford University Center for Fundamental Model Research (CRFM), aimed at systematically evaluating large language models through multidimensional, standardized, and reproducible

2026-05-30

A Quick Take: Just how much hassle can that AI meeting note-taker hidden inside Baidu Netdisk really save you?

144

"Jiandan Tingji" is an AI-powered speech-to-text tool launched by Baidu Netdisk. Integrated with the ERNIE Bot (Wenxin Yiyan) large language model, it supports the transcription of meetings, interviews, and lectures, as well as the intelligent generation of meeting minutes, achieving an accuracy rate of up to 97%. It is available across all platforms, with a continuous monthly subscription priced at 25 yuan.

2026-06-29

Alibaba's M6—the world's first 10 trillion-parameter multimodal large model—consumes only 1% of the energy of GPT-3.

135

M6 is a general-purpose multimodal large model developed by Alibaba DAMO Academy. In October 2021, it became the world's first AI pre-trained model with 10 trillion parameters. Based on the self-developed Whale distributed framework, it was trained in just 10 days using only 512 V100 GPUs, with energy consumption only 1% of GPT-3. It supports tasks such as image and text generation, visual question answering, and poetry creation, and has been deployed in more than 40 scenarios, including Tmall virtual anchors and Taobao search. It is the core predecessor of the Tongyi Thousand Questions large model.

StableLM—an open-source AI model from the same family as Stable Diffusion, capable of running with only 3B parameters and commercially viable.

133

StableLM is an open-source large language model series launched by Stability AI. It is trained on The Pile extended dataset, which contains 1.5 trillion tokens. It offers multiple parameter versions, including 1.6B, 3B, 7B, and 12B, and supports text generation and code writing. It is licensed under the CC BY-SA 4.0 open-source license, allowing free commercial use. Stable LM 2 12B outperforms Llama 2 70B in some benchmark tests. It is suitable for developers, researchers, and small and medium-sized enterprises for private deployment.

SuperCLUE—A benchmark for Chinese large models: from foundational capabilities to AI agents, a single test reveals who is "swimming naked."

129

SuperCLUE is a comprehensive evaluation benchmark for general-purpose Chinese large models, released by the CLUE team. It provides authoritative, multi-dimensional capability assessments and rankings for Chinese large models by regularly publishing monthly and semi-annual reports based on three key benchmarks—open-domain multi-turn dialogue, closed-domain objective questions, and anonymous head-to-head battles—as well as dimensions such as mathematical reasoning, code generation, and AI agents.

2026-07-10

LLaMA—Meta's open-source AI model revolution, with performance comparable to GPT-4 and completely free for commercial use.

122

LLaMA is an open-source family of large language models launched by Meta, evolving from LLaMA 1 in February 2023 to multimodal LLaMA 4. The 70B parameter version boasts performance close to GPT-4, while the 405B version rivals GPT-4o. It is completely open-source and commercially viable. Enterprises can download and deploy it on their own servers without data sharing, meeting compliance requirements in industries such as finance and healthcare. With over 1.2 billion downloads globally, it is suitable for developers and enterprises requiring private deployment and deep customization.

PubMedQA—an "AI benchmark" specifically designed for biomedical question answering; only by comprehending research papers can an AI truly pass the "Medical Turing Test."

119

PubMedQA is the first question-answering dataset requiring reasoning over biomedical research texts; it was released in 2019 by institutions including the University of Pittsburgh. The task involves answering "Yes," "No," or "Maybe" questions based on PubMed abstracts. Comprising 1,000 expert-annotated samples and 211,000 artificially generated ones, the dataset aims to evaluate the ability of AI models to comprehend and reason about complex medical literature. It is widely used to benchmark the performance of large language models in the medical domain.

2026-07-10

Vocu AI — A domestic large-scale speech model ranked #1 globally on the HuggingFace TTS Arena, capable of infusing AI-generated speech with a human touch and genuine emotion.

114

Vocu AI is an AI speech synthesis platform developed in-house by Guangzhou Shuogu Technology; its V3 series models ranked first globally in the Hugging Face TTS Arena blind tests. It offers both instant cloning (requiring a 3-second sample with over 95% similarity) and professional-grade cloning. Supporting more than 30 languages and dialects, the platform delivers precise emotional expression and cinematic-quality performance. API access is available to facilitate efficient integration for developers.

2026-07-11

WenXiaoBai - Proactive AI Assistant with Deepseek R1/V3/V3.1, AI Web Search, Academic Search, Image Editing & Emotional Companion

107

WenXiaoBai is a proactive AI assistant designed to be your AI companion for exploring the world. It offers top-tier large models for free (including Deepseek R1/V3/V3.1, and WenXiaoBai 5 comparable to OpenAI GPT-5), supporting AI web search, AI academic search, Deep Research, AI image editing and generation, and AI agent emotional companionship. Its Pro long-form content delivery enables long-form writing, theses, reports, proposals, deep research, official documents, and PPT generation, allowing users to upload images, documents, PPTs, tables, text, and other file types for interaction.

2026-07-22

MiTa AI Search – a free AI search engine that thinks and provides references, so you no longer need to flip through hundreds of pages of materials when doing research.

107

MetaAI Search is an intelligent search engine launched by Shanghai MetaNet Technology Co., Ltd. based on its self-developed MetaLLM large-scale model. It offers three search modes: concise, in-depth, and research-oriented. It supports searches across the entire web, document libraries, academic resources, podcasts, and more. Search results automatically generate outlines and mind maps, with each result accompanied by references. It incorporates the DeepSeek R1 deep thinking model, supporting a "think first, then search" research mode. It is compatible with web pages, apps, and mini-programs, with a free daily search limit of 100 times. Suitable for students, researchers, and professionals, it transforms information retrieval from "providing links" to "providing direct answers."

BaiXiaoYi - Baichuan Intelligent Medical AI Assistant for Online Consultation and Health Advice

104

BaiXiaoYi is a medical and health AI assistant developed by Baichuan Intelligent, leveraging large language model technology to provide online consultation, symptom analysis, medication advice, and health knowledge queries. It simulates the doctor consultation process, integrating medical knowledge bases to deliver professional and caring health guidance, helping users understand their conditions and obtain medical recommendations. BaiXiaoYi aims to make reliable medical information accessible to everyone anytime, reducing healthcare resource pressure and improving health management efficiency.

2026-07-22

360AI Search – an AI search engine that can "think slowly" for you, so you no longer need to flip through dozens of pages to find information.

103

360AI Search (Nano AI Search) is an AI-native search engine launched by 360, supporting multimodal input including text, voice, photos, and videos. It features a built-in "slow thinking mode," capable of calling up three or more large models at once and generating a 5,000-word in-depth report after reading 250,000 documents. It supports extracting mind maps from 100-page PDF/Word documents in one minute, extracting key points from one hour of video in one minute, and providing a one-stop AI tool for webpage AI summarization, PPT generation, and academic source tracing. In July 2024, user visits exceeded 90 million, transforming search from "providing links" to "providing direct answers."

ZCOOL – China's inspiration garden for designers, where 18 million creatives find direction.

ZCOOL is China's largest interactive community for designers. Founded in 2006, it brings together 18 million designers, illustrators, photographers and other creative professionals.

2026-07-04

Rabbit R1 – An AI hardware device that challenges the "App terminator," a small orange box that went from being ridiculed to gaining popularity.

The Rabbit R1 is an AI-native pocket assistant launched by Rabbit Inc. It features a Large Motion Model (LAM) and allows users to directly control apps and complete tasks via voice commands. Although it was controversial due to its incomplete functionality after its release in early 2024, its reputation has reversed after more than 35 OTA updates over two years and a rewrite of Rabbit OS 2. The company also plans to launch a more geeky Cyberdeck product.

2026-07-09

360 Smart Drawing – an AI drawing tool equipped with a massive model containing hundreds of billions of parameters, capable of generating thousands of styles with a single click.

360 Smart Drawing is an AI image creation platform launched by 360 based on its self-developed "360 Brain" large-scale model with hundreds of billions of parameters. Integrating the 360CV large-scale model and multimodal technology, it provides full-scene AI drawing and retouching functions, including text-to-image, image-to-image, doodle-to-image, partial redrawing, AI portrait, and LoRA model training. It has built-in models for thousands of styles, including traditional Chinese style, realistic, animation, and 3D, and supports bilingual (Chinese and English) prompts. As a member of the 360 AI ecosystem, it has been integrated into the entire 360 platform ecosystem.

2026-07-04

Cherry Studio – an all-in-one AI workstation, a desktop client that provides unified access to 300+ mainstream large-scale models.

Cherry Studio is a cross-platform AI desktop client that integrates over 300 mainstream models from OpenAI, Claude, Gemini, and other sources, as well as local Ollam models. It supports intelligent dialogue, autonomous agents, document processing, global search, and other functions, with local data storage for enhanced privacy and security. The community edition is completely free and open-source, supporting Windows, macOS, and Linux.

Beijing Academy of Artificial Intelligence - Leading AI Research Institute in China, Driving Cutting-Edge AI Technology and Open-Source Ecosystem

The Beijing Academy of Artificial Intelligence (BAAI) is a leading AI research institute in China, focusing on cutting-edge AI technology development, open-source ecosystem building, and talent cultivation. Core features include: releasing world-leading AI large models (e.g., WuDao series), building open-source platforms (e.g., FlagOpen), hosting AI academic conferences (e.g., BAAI Conference), and promoting AI ethics and safety research. It serves researchers, developers, enterprises, and policymakers, providing comprehensive support from fundamental research to industrial application.

2026-07-15

ModelScope – China's largest and most active open-source AI model platform, boasting a "model-as-a-service" ecosystem with over 30 million developers.

ModelScope is a leading open-source AI model platform in China, initiated by Alibaba DAMO Academy in conjunction with the CCF Open Source Development Committee. Adhering to the "Model as a Service" (MaaS) concept, the community has gathered over 200,000 open-source models, 30 million developers, and over 500 contributing organizations. It provides end-to-end services from model experience, download, fine-tuning, training to deployment, covering LLM, multimodal, speech, and AIGC fields. As a core infrastructure for AI developers in China, it is promoting the democratization of AI technology through "open source."

MiniMax Audio—an ultra-realistic large-scale speech model capable of everything from 10-second voice cloning to support for over 40 languages, enabling AI to speak with the warmth of a real person.

MiniMax Audio is an AI voice platform under MiniMax that offers features such as text-to-speech, voice cloning, and music generation. Its proprietary Speech series models support over 40 languages, ultra-long text, emotional expression, and low-latency interaction, and have been adopted by leading global platforms and products such as LiveKit, Pipecat, Gaotu, and Ximalaya.

2026-07-11

Moyin (Moyin Workshop)—an AI voiceover powerhouse featuring over 800 voices and 1,000 styles, trusted by creators of short videos and audiobooks.

Moyin (Moyin Workshop) is an AI voice synthesis platform under Mobvoi. Powered by the proprietary "Sequence Monkey" (Xulie Houzi) large model and a fifth-generation TTS engine, it offers over 800 voice profiles, more than 1,000 styles, and nearly 20 fine-tuning features. The platform supports multiple languages and dialects, voice cloning, cloud-based video editing, and multi-user collaboration; it is widely used in applications such as short videos, audiobooks, and film/TV commentary, having served over 6 million users to date.

2026-07-11

Exa—a search engine infrastructure built specifically for AI agents, enabling large models to truly understand the web.

Exa is an infrastructure company providing search APIs for AI agents. Unlike traditional search engines, it uses neurosemantic search technology to enable AI to truly "understand" query intent, rather than simply matching keywords. Its self-built search engine covers over 500 billion URLs, offering multi-layered APIs ranging from ultra-fast searches (less than 200 milliseconds) to in-depth research. It has been adopted by over 5,000 companies, including Cursor, Cognition, and HubSpot, and by 400,000 developers. In 2026, it completed a $250 million Series C funding round, valuing the company at $2.2 billion.

GPT-4o—OpenAI's all-around multimodal flagship model, featuring real-time voice and video interaction, and freely available to all users.

GPT-4o (“o” stands for Omni) is OpenAI's next-generation flagship multimodal large model, released in May 2024. It enables real-time inference across text, audio, images, and video, with an audio response time as fast as 232 milliseconds, approaching human conversational reaction speed. It possesses groundbreaking capabilities such as emotion perception, real-time translation, and image generation. Its API is twice as fast as GPT-4 Turbo, costs only half the price, and is available to all free ChatGPT users.

Runway—a world-leading AI video generation and creative tool platform—makes imagination instantly visible, from text to cinematic visuals.

Runway is a leading global AI video generation and creative tools platform, providing designers, filmmakers, and creative professionals with a complete AI creation suite ranging from text-based and image-based video to precise camera control. Its Gen-4.5 model is the world's highest-rated video model, supporting the creation of cinematic, highly realistic video content. The platform also offers cutting-edge features such as Runway MCP, real-time video character generation, and video editing, boasting a large enterprise customer base and reshaping the way visual content is created.

There are 54 more tools not shown. Try more specific keywords.

AI News (13)

Meituan Open-Sources LongCat-2.0, a 1.6-Trillion-Parameter MoE Model; Full-Stack Execution on Domestic Computing Hardware Successfully Achieved

658

Meituan has officially open-sourced LongCat-2.0, a massive MoE model with 1.6 trillion parameters. This release achieves end-to-end training and inference deployment on a domestic computing cluster comprising 50,000 GPUs, thereby breaking reliance on overseas computing power. The announcement details LongCat-2.0's technical highlights, performance advantages, and industry value.

OpenAI's GPT-5.6 series preview remains in internal testing, with proprietary inference chips set for mass production by year-end.

277

OpenAI is officially advancing the limited beta testing of its GPT-5.6 model series and has announced a partnership with Broadcom to co-develop "Jalapeño," a dedicated inference ASIC. The project targets tape-out within nine months and large-scale deployment by the end of 2026, aiming to slash inference costs by 50% and effectively completing the company's vertically integrated strategy combining models and chips.

Anthropic's valuation has reached $965 billion, surpassing OpenAI for the first time.

260

The AI industry is undergoing a seismic shift as Anthropic’s post-investment valuation climbs to $965 billion, surpassing OpenAI for the first time. Its Claude Opus 4.8 model has reached the pinnacle of reasoning capabilities, while Claude Code captures over half the code-focused AI market with $6.3 billion in annual revenue; this analysis delves into the company's three core strengths: capital, technology, and commercialization.

The wave of mandatory retirement is coming! Codex launches GPT-5.2/5.3-Code in June, developers

211

Sun Zhengyi was the richest man in Asia again, and fell again three days later: how to understand

161

To be honest, my first reaction when I saw this news was not "wow", but "is it again?" - because the last time he was so successful was in 2017. After that, everyone knows the story: WeWork went bankrupt, SoftBank Vision Fund suffered huge losses, his personal wealth once shrank to $21.1 billion,

How do you view Minimax's new M3 multimodal model and updated Token Plan?

150

I originally wanted to say 'M3 is very strong' or 'just like that' directly, but later I found out that it's not something that can be explained in a single sentence. Let me admit one thing first: I have a bias against Minimax. It's not the kind of malice, it's the prejudice of being 'tricked'.

The independent variable robot releases the world's first "event level prediction" embodied

129

Breaking the limitations of traditional frame by frame learning, the way robots understand tasks has entered a new stage

Out of control and overturned! Google AI unauthorized tampering with production code, resulting in

104

An In-Depth Review by the Former COO of Zhipu AI: Deconstructing Five Common Pitfalls in Enterprise AI Implementation and Breaking the Deadlock of Stalled Deployment

Zhang Fan, former COO of Zhipu AI and founder of Yuanli AI, draws on extensive experience serving government and enterprise clients to identify five common pitfalls in corporate AI implementation: impressive demos that fail upon official launch; "bottomless pit" fully customized deliveries that derail projects; project initiation based on subjective, unrealistic expectations of returns; a focus on technical deployment while neglecting business process re-engineering; and AI projects getting stuck in pilot phases without achieving large-scale rollout. These issues stem from three fundamental misconceptions: general-purpose large models do not equate to enterprise business capabilities; technical feasibility does not guarantee commercial returns; and success in a localized pilot does not ensure replicability across the entire organization. To address this, he proposes a progressive implementation roadmap—moving from conversational assistants and business support tools to automated workflows, AI agents, and finally, business operating systems—emphasizing the need to refine reusable business templates before scaling up to a company-wide rollout.

Hands-On Guide to Large Models: 5 Real-World Cases to Get You Started Fast, with Common Problem Solutions

Practical Guide to Large Model Development: A Real Path from Confusion to Proficiency Folks, have you been bombarded by the term "large models" lately to the point of questioning everything? 🤯 From C...

AI Model Comparison 2026: Top LLMs Benchmarked vs Latest Models - Who Leads?

Introduction: In 2026, the "Clash of AI Titans" Has Reached a Fever Pitch Folks, if you're still stuck thinking "Is ChatGPT the best?", you really need to catch up. The AI large model race in 2026 is ...

LLM Comparison Deep Dive: Technical Architecture, Capability Benchmarks, and Use Case Analysis

Preface: The Age of AI Models — Which One is Right for You? Hey folks, are you feeling overwhelmed by the constant stream of news about large language models? From ChatGPT to Ernie Bot, from Claude t...

AI Model Comparison 2026: Performance, Cost & Use Cases Reviewed