Assistant AI apps

Zhipu's GLM assistant — frontier open-weight chat, free to use.

Z.ai is the international assistant of Zhipu AI, built on the GLM model family. The free chat runs the flagship GLM-5 and GLM-5.1 models with reasoning and agentic modes, and the same models are served to developers over a low-cost, OpenAI-compatible API. GLM-5's open weights ship under the MIT license.

Perplexity's AI browser with a sidebar assistant that acts across your tabs.

An AI-native Chromium browser from Perplexity that puts a persistent assistant in the sidebar — it summarizes pages, compares products, and takes actions like drafting emails or managing tabs using the context of what you're browsing. Perplexity search is the default, with Deep Research and voice mode built in. Free on Windows, macOS, Android, and iOS; an optional $5/month Comet Plus tier unlocks premium publisher content.

StepFun's multimodal assistant and Step-model API platform.

The chat assistant and developer platform of StepFun, the Shanghai AI lab behind the Step model family. The assistant at stepfun.ai bundles web search, knowledge-base Q&A, image generation, and Step-Audio speech tools, while the open platform at platform.stepfun.ai serves Step language, audio, and image-edit models over an international API.

OpenAI's web browser with ChatGPT built in — sidebar, memories, and agent mode.

OpenAI's Chromium-based web browser with ChatGPT woven through it: a sidebar that can summarize, compare, and answer questions about any page, optional browser memories that carry context across sessions, and an agent mode that navigates sites and completes multi-step tasks on your behalf. It launched on macOS in October 2025, with Windows, iOS, and Android versions announced. Free with any ChatGPT account; agent mode requires a paid plan.

An AI-native web browser with chat and skills built into the URL bar.

An AI-native desktop web browser from The Browser Company, the team behind Arc. Dia bakes a chat assistant and reusable 'skills' directly into the address bar, with awareness of your open tabs and browsing history, so you never need to copy context into a separate chatbot. It launched publicly on macOS in 2025. A free tier covers light AI use; a $20/month Pro plan removes the AI usage limits.

Generate presentations, websites, and documents from a single prompt.

An AI design tool that turns a prompt or pasted content into polished presentations, websites, documents, and social posts — automatically handling layout, copy structure, and visuals. It can import and restructure existing PDFs, slide decks, and Word files, and supports interactive embeds like live charts, videos, and forms. A Generate API added in early 2026 lets teams produce content programmatically at scale.

Full-stack enterprise generative AI platform for building agents and apps.

An enterprise platform that pairs an application layer for building AI agents and workflows with its own foundation models, the Palmyra LLM family. It targets regulated industries with SOC 2 Type II, PCI, and HIPAA compliance and a no-training-on-customer-data policy. Specialized models like Palmyra Med and Palmyra Fin tune the platform for healthcare and finance use cases.

Productivity launcher with built-in AI and a 1,500+ extension store.

A keyboard-driven command launcher that replaces Spotlight with app launching, clipboard history, window management, snippets, and a large store of community extensions. Raycast AI adds a model-switching assistant and quick AI commands directly in the command bar. Native on macOS with a Windows version in beta as of late 2025.

One app for many AI models — chat and build bots across providers.

Quora's aggregator that puts models from Anthropic, OpenAI, Google, xAI, Meta, and others behind a single subscription, alongside image and video bots. Anyone can build a custom no-code bot in minutes, and creators are paid based on usage. Available on web, mobile, and desktop, with an API for developers.

Work AI: enterprise search, assistant, and agents over your apps.

A workplace AI platform that unifies search, an assistant, and agents across 100+ connected enterprise apps (Slack, Teams, GitHub, ServiceNow and more). It enforces each user's existing access permissions over the indexed content. Sold enterprise-only — pricing is quote-based via sales, with no public self-serve tier.

Moonshot's assistant — long-context chat, deep research, and agents.

Moonshot AI's assistant — long-context chat, Deep Research, and an agent mode (Kimi Code, parallel subagents). Built on the open-weight Kimi K2 model family.

Anthropic's assistant. Thoughtful, long-context, strong at code + writing.

The Claude consumer app — chat with Projects, artifacts, file analysis, and MCP connectors. Known for careful reasoning, large context, and being a strong writing + coding partner.

Open-source ChatGPT alternative that runs 100% offline on your computer.

An open-source desktop assistant that runs AI models 100% offline on your own machine, bundling a local model runner so you can download and chat with open models like Llama, Gemma, and Qwen. It also supports bring-your-own keys for cloud providers when you want them, and exposes a local OpenAI-compatible server. Fully open source under Apache 2.0 with no paid tier.

Local-first personal AI assistant you run on your own devices, on any platform.

A self-hosted, local-first personal AI assistant with a gateway as the control plane, reachable across messaging channels like WhatsApp, Telegram, Slack, and Discord. It runs on macOS, Linux, Windows, iOS, and Android, with an onboarding wizard for setup. Free and open source under MIT; you supply your own model API key.

Mistral's fast, European assistant with web search and code.

Mistral AI's assistant — chat with web search, document analysis, image generation, and code. Fast, privacy-minded, and built on Mistral's own open and frontier models.

The default AI assistant. Chat, voice, vision, and a tool ecosystem.

OpenAI's consumer assistant — multimodal chat with voice, image input, web browsing, code, and custom GPTs. The product that put generative AI in everyone's hands.

Microsoft's assistant across Windows, Edge, and Microsoft 365.

Microsoft's everyday assistant — chat, image generation, and Copilot Vision, woven into Windows, Edge, and the Microsoft 365 apps. The consumer front-door to Microsoft's AI.

Self-hosted, open-source ChatGPT alternative unifying every major AI provider.

A self-hosted AI chat platform that puts many providers behind one ChatGPT-style UI, with agents, code interpreter, web search, image generation, artifacts, file analysis, and multi-user auth. You run it yourself via Docker and supply your own model keys, or point it at a local Ollama/OpenAI-compatible endpoint. Fully open source under MIT — no cost for the software itself.

xAI's assistant with real-time X access and a less-filtered voice.

xAI's assistant — chat, image generation, and real-time knowledge pulled from X. Positioned as witty and current, with deep integration into the X platform.

Open-source, self-hosted AI chat and enterprise search over your own docs.

Onyx (formerly Danswer) is an open-source AI chat and RAG platform that connects to your company's docs and apps for grounded, cited answers, and works with any LLM. It is self-hosted via Docker/Kubernetes and supports local models, keeping data on your own infrastructure. The core is MIT-licensed and free; an open-core model puts optional enterprise features under a separate license, and the vendor also offers a managed cloud.

Google's assistant, wired into Search, Workspace, and Android.

Google's multimodal assistant — chat, image generation, deep research, and tight integration across Gmail, Docs, and Android. Backed by the Gemini model family.

Multi-model AI workspace with chat, voice, image gen, and tool use — BYO keys.

A self-hostable AI workspace for chatting across hundreds of models from many providers, with voice calls, image generation, web search, and code execution. It runs self-hosted via Docker or Vercel and connects to local engines (Ollama, LM Studio, LocalAI) as well as cloud providers. MIT-licensed and free — you supply your own provider API keys.

Self-hosted AI workspace with a ChatGPT-style UI for local and API models.

A self-hosted, privacy-first AI workspace offering a ChatGPT/Claude-like interface with agent capabilities, memory, and productivity tools (PDF/Office, built-in MCP servers). It connects to local models via llama.cpp, Ollama, or vLLM, or to API providers like OpenAI and OpenRouter, and ships Docker support for Linux, macOS, and Windows. Free and open source under MIT.

Desktop AI client unifying frontier LLMs, local models, and 300+ assistants.

A cross-platform desktop AI productivity app for chatting across cloud and local models, with autonomous agents and 300+ prebuilt assistants. It runs on your own machine and supports local models via Ollama and LM Studio alongside cloud providers using your own keys. The community edition is AGPL-3.0; organizations over ten people require a paid commercial license.

All-in-one private AI app for chatting with your documents, with agents.

An all-in-one application for private, ChatGPT-style chat over your own documents, with built-in RAG, AI agents, and multi-user workspaces. Runs as a local desktop app (Mac/Windows/Linux) or self-hosted via Docker, and supports 40+ LLM providers plus local models with your own keys. Open source under MIT; Mintplex Labs also offers a paid hosted instance, making it freemium.

Open, low-cost chat with strong reasoning. Free to use.

DeepSeek's assistant — chat with a reasoning mode and web search, backed by the open-weight DeepSeek models that reset the cost curve for frontier-grade quality.

Open-source personal AI assistant built on tools, memory, and agentic execution.

A self-hosted personal AI assistant organized around tools, context, memory, and agentic execution, with smart, controlled, and agent run modes. It runs on your own machine and can use local models and local context instead of routing everything through third-party services. MIT-licensed and free; currently shipping a 2.0 developer preview alongside the legacy stable branch.