Lumina — 400+ Curated AI & Developer Tools, Free Discovery Platform

Claude AI is a family of large language models (LLMs) and a conversational AI chatbot developed by the research firm Anthropic. It is known for its focus on AI safety and ethics, and its capabilities include natural language processing, generating text and code, answering questions, and analyzing both text and visual input.

Gemini can refer to Google's family of multimodal large language models (LLMs) or a version of the LLM that powers a chatbot and an assistant, previously known as Bard. It can also refer to the third astrological sign, representing the twins. The Google Gemini LLM is a competitor to OpenAI's GPT models and is used across various Google products, from the Gemini chatbot to features within Google Workspace and Android devices.

Google AI Studio is Google's free web platform for prototyping and chatting with the latest Gemini models, including Gemini Flash and Pro variants. It offers generous free limits with a Google account, along with system instructions, model tuning, and API key generation for developers who want to build on top of Gemini.

Qwen is a family of large language models (LLMs) and an AI assistant developed by Alibaba Cloud. It is known for being open-source and having strong capabilities in text and code generation, multimodal tasks (like image and video processing), and multilingual support. The name is a Chinese acronym for "to comprehend meaning, and answer a thousand questions".

DeepSeek is a Chinese artificial intelligence research lab that develops open-source large language models (LLMs). Founded in 2023 by Liang Wenfeng, it gained significant attention in early 2025 for its powerful yet cost-effective models like DeepSeek-V3 and DeepSeek-R1, which use a Mixture-of-Experts (MoE) architecture. Its open-source nature allows anyone to inspect, modify, and run the models, promoting transparency and innovation in the AI field.

Meta AI is an AI assistant that functions as a virtual assistant and chatbot, integrated into Meta's family of apps like WhatsApp, Instagram, and Facebook, and also available as a standalone web experience. It uses generative AI to answer questions, help with tasks like image generation, and provide information, often by searching across Meta's apps. Users can interact with it through direct chats or by tagging "@Meta AI" in group conversations.

Grok is an AI chatbot developed by Elon Musk's company, xAI, designed to have a more sarcastic and "rebellious" personality than other assistants. It integrates with the social media platform X (formerly Twitter) to access real-time information and is intended to provide witty and opinionated responses to controversial topics. The technology is named after the concept of profoundly understanding something, a term popularized by Robert Heinlein's novel Stranger in a Strange Land.

ChatGPT is an advanced AI chatbot developed by OpenAI that uses a large language model to hold human-like conversations and generate responses to user prompts. It can write various types of content, answer questions, debug code, and more, by predicting the most probable next word based on its vast training data.

Microsoft Copilot is an AI assistant designed to boost productivity by providing intelligent suggestions, automating tasks, and helping with various content creation and analysis.

Mistral AI is a French AI startup that develops large language models (LLMs) and other generative AI technologies, competing with major AI companies like OpenAI and Google. Founded in 2023, the company focuses on creating powerful, efficient, and often open-source models to make advanced AI more accessible to businesses and developers. Their models are used for various tasks such as chatbots, content creation, coding, and complex document analysis.

Kimi is an AI chatbot and large language model developed by Moonshot AI. Its most recent version, Kimi K2 Thinking, is an open-weight model known for its advanced "agentic" capabilities, allowing it to perform complex tasks independently by planning and using tools over long periods without human intervention. It is recognized for strong performance on benchmarks, a cost-effective price, and a Mixture-of-Experts (MoE) architecture.

HuggingChat is a free, open-source AI chatbot developed by Hugging Face that serves as an alternative to proprietary models like ChatGPT. It allows users to chat with various open-source AI models, such as Llama and Mistral, and can generate text, answer questions, and even browse the web to find information.

Pi (short for "Personal Intelligence") is a conversational AI chatbot developed by Inflection AI, designed to be an empathetic, supportive, and highly personal assistant. Unlike task-oriented bots, Pi focuses on dialogue, offering advice, brainstorming, and acting as a sounding board in a friendly, human-like manner.

Duck.ai, also known as DuckDuckGo AI Chat, is a free and private service that gives users access to various AI chatbots for tasks like answering questions, generating text, and summarizing information. It is a feature of the privacy-focused search engine DuckDuckGo and is designed to protect user privacy by anonymizing conversations and not using them for AI model training. You can access it through the DuckDuckGo search interface or a dedicated chat link.

Lumo is a privacy-focused AI assistant from Proton, the makers of Proton Mail, powered by open-weight models like Qwen and GLM. It features zero-retention encrypted conversations that are never used for training, along with file analysis, web search, and saved-chat history protected by end-to-end encryption.

NVIDIA AI Foundation Models (build.nvidia.com) is an interactive platform for developers to explore, test, and deploy NVIDIA-optimized generative AI models. Access pre-trained models for text, image, and code generation, with API integration for building AI-powered applications.

Together.ai is a multi-model chat platform offering free daily access to leading open-weight models like Kimi, Qwen, and DeepSeek in one interface. With up to 110 requests per day and easy model switching, it is one of the simplest ways to compare frontier open models side by side.

Z.ai is an AI platform and conversational chatbot developed by the Chinese company Zhipu AI. It is powered by the company's GLM series of large language models and is designed for tasks like coding, reasoning, and creative content generation. Z.ai is noted for its agentic capabilities, which allow it to break down complex tasks into smaller sub-tasks, and for features like a "Full Stack Coder" and the ability to generate AI slides.

Poolside is an AI lab building open-weight foundation models for software engineering. Its chat interface lets developers work with the Laguna family of agentic coding models — including Laguna S 2.1 (118B parameters, 1M context) and Laguna XS 2.1 (33B, small enough to run on-device) — for code generation, debugging, and multi-step engineering tasks, with desktop and CLI apps plus availability through OpenRouter.

ERNIE Bot (Enhanced Representation through Knowledge Integration) is an advanced conversational AI chatbot developed by Chinese tech giant Baidu. Released in 2023, it is a multimodal model (ERNIE 4.0/4.5/5.0) that understands and generates text, images, audio, and video, competing with GPT-4 in Chinese and English natural language tasks.

MiniMax AI refers to a suite of advanced artificial intelligence tools from the Chinese AI company MiniMax, known for its powerful text-to-video generator (Hailuo AI) that creates cinematic videos from prompts, but also encompasses AI agents for complex tasks like website building, music generation, and large language model capabilities (like Hailuo 2.3), aiming to be an all-in-one platform for creators and developers.

DeepAI.org is an all-in-one creative AI platform that provides tools for users to generate images, videos, and music from text prompts, edit photos, and chat with AI. Founded in 2016, it aims to make AI accessible to everyone, from individuals to large businesses, offering a range of features through a web browser and APIs. It includes both free and paid versions of its services.

Falcon LLM is a suite of open-source large language models developed by the Technology Innovation Institute (TII) in Abu Dhabi. These models are known for their high performance, efficiency, and accessibility through open-source licenses, allowing developers and companies to use them freely in applications and services. Falcon LLMs come in various sizes, from smaller 1.3 billion parameter models to much larger 180 billion parameter versions, and support a range of tasks, including text, vision, video, and audio understanding.

Allen Institute for AI (AI2) is a non-profit research institute founded by Microsoft co-founder Paul Allen, dedicated to building breakthrough AI for the common good, focusing on open models (like Olmo), scientific discovery (Asta), and planetary problems through research in NLP, vision, robotics, and common sense AI, aiming to make advanced AI accessible.

Reka AI is a research-driven AI company that builds multimodal models for understanding text, images, video, and audio. Known for efficient, deployable models like Reka Core, they specialize in enterprise solutions, including Reka Vision for surveillance, and Reka Clip for automated video editing.

Solar Pro by Upstage AI is an advanced, high-performance large language model (LLM) designed to operate efficiently on a single GPU. With 22 billion parameters (and an updated 31B version, Solar Pro 2), it delivers performance comparable to much larger 70B+ models, specializing in structured text understanding, complex reasoning, and multilingual tasks for enterprise applications.

Arcee AI is a platform for deploying and interacting with domain-specific small language models (SLMs). It offers models at various sizes — Trinity-Nano (6B), Trinity-Mini (26B), and Trinity-Large (400B) — for different deployment scenarios from local consumer GPUs to cloud endpoints. Trinity-Large is a robust generalist with strong performance across reasoning, coding, math, and complex task decomposition.

Inception from Inception Labs is an AI research company developing "Mercury," the world's first commercial-scale diffusion-based large language model (dLLM). Unlike traditional auto-regressive models that generate text token-by-token, Inception's diffusion models generate text in parallel, creating significantly faster (up to 10x) and cheaper AI with superior control over output.

StepFun is a Shanghai-based AI startup founded in 2023 by former Microsoft employees, focused on developing multimodal large models (LLMs) and artificial general intelligence (AGI). It specializes in high-quality text-to-video, image, and audio generation, often offering open-source models. Their technology enables advanced reasoning and creative content generation.

LongCat-Flash-Chat is an open-source, 560-billion parameter Mixture-of-Experts (MoE) AI model developed by Meituan (released Sep 2025), designed for high-performance, efficient, and cost-effective conversational and agentic tasks. It features 128K context length, specializing in reasoning and tool-use, while operating at roughly 27B active parameters.

Apertus is a Swiss open-source, multilingual large language model (LLM) developed by ETH Zurich, EPFL, and the Swiss National Supercomputing Centre (CSCS). Released in September 2025, it aims to be a transparent and sovereign AI, complying with European AI regulations while being trained on 15 trillion tokens to support over 1000 languages.

Stack Overflow AI, also known as stackoverflow.ai, is an AI-powered tool developed by Stack Overflow to provide instant answers and educational guidance for developers and technologists. It aims to combine the vast, community-verified knowledge base of Stack Overflow with the capabilities of a large language model (LLM).

Tencent Hy3 (HYookie3) is an advanced image generation and editing AI space developed by Tencent. It provides powerful text-to-image generation, image editing capabilities, and multimodal generation features through an intuitive interface on Hugging Face Spaces.

Tencent AI Studio is a cloud-based AI development platform by Tencent that provides online IDEs, GPU resources, and pre-trained models for training, fine-tuning, and deploying AI models. It offers a collaborative environment with Jupyter notebooks, model hosting, and community features for AI developers and researchers.

Xiaomi AI Studio is an AI platform developed by Xiaomi that provides access to their MiMo multimodal large language model. It offers AI chat capabilities, image generation, and other AI-powered tools for developers and consumers.

Free AI platform gives you access to 1000+ free AI tools powered by GPT-5.2, Claude-Sonnet-4.5, Gemini-3-Pro, DeepSeek-R1. Generate AI images powered by Nano Banana, Z-AI, Hidreams, Qwen and imagen-4. Create AI contents, AI Voiceovers, AI Stickers and chat with AI for free

Sarvam is India's full-stack sovereign AI platform with frontier-class LLMs (Sarvam-30B, Sarvam-105B) trained from scratch under the IndiaAI Mission. Supports speech, text, and translation APIs across all 22 official Indian languages. Open source (Apache 2.0) with a chat playground at dashboard.sarvam.ai.

Indic LLM Arena is a crowd-sourced, human-in-the-loop platform by AI4Bharat (IIT Madras) for benchmarking LLMs on Indian languages, context, and safety. Chat with models, compare responses side-by-side in blind tests, and vote to help rank models via statistically robust ELO scores. Supports voice and text input across Indian languages including code-mixed variants like Hinglish.