Together AI
Together AI is a full-stack cloud platform for running, fine-tuning, and deploying open-source AI models, aimed at engineering teams that need production inference without managing their own hardware.

Jan is an open-source desktop chat application that lets users run open models on their own hardware or connect to hosted providers like OpenAI, Anthropic, and Google. It targets privacy-conscious individuals and teams who want a friendlier local LLM experience than command-line tooling offers.
Jan is an open-source desktop chat application for running AI models on your own machine or connecting to hosted providers, aimed at people who want a private, approachable alternative to browser-based chatbots. It is developed by Menlo Research and distributed through GitHub, with a Linux build available as a Flatpak on Flathub.
Jan's pitch rests on usability as much as openness. Users who have tried LM Studio and Ollama describe those tools as capable but engineer-built, while praising Jan for a simpler, more polished interface. That framing matters: the app is meant to make local language models feel like a normal desktop application rather than a developer utility. The project has drawn 7.2M+ downloads and a Discord community of more than 15,000 members, alongside roughly 45,000 GitHub stars.
The model picker is the core of the product. Jan can run open models directly on the user's hardware, and it can also route requests to hosted providers. The site lists ChatGPT from OpenAI, Claude from Anthropic, Gemini from Google, Llama from Meta, Mistral, Qwen from Alibaba, DeepSeek, Gemma from Google, and Kimi from Moonshot AI. That mix lets a user keep a private local model for sensitive work and switch to a hosted model when a task needs more capability, all without leaving the same chat window.
Jan integrates with Hugging Face, so users with a Hugging Face PRO account can pull trending models into the app by pasting a model ID under Settings, then Providers. Users have also reported running web search from within Jan before generating a response, which brings a retrieval step to local conversations. The app additionally supports MCP options, giving technically inclined users a way to extend what the assistant can reach.
The project separates Jan Desktop from Jan Agent. According to the team's handbook, Jan Agent is a general-purpose agent that takes a task, such as research for a brief or an analysis, and works on it for as long as needed, whether that is five hours or ten. The framing is deliberately practical: hand off the work, let the agent run, and move on. Documentation for both products lives at docs.jan.ai.
A memory feature is listed as coming soon. The concept shown on the site is a persistent profile that remembers a user's preferences and context, so answers stay consistent without repeated setup. Until it ships, users should expect each conversation to start closer to a blank slate. That is a meaningful gap for anyone comparing Jan to assistants that already carry context between sessions.
Jan makes the most sense for people who care about where their prompts go. Running a model locally means conversation data stays on the machine, and the open-source code can be inspected or self-hosted. The tradeoff is that local performance depends on the hardware, and cloud connections require the user's own provider accounts. For users who want a clean desktop front end to both worlds, Jan offers a credible option. The project is built in public, and its roadmap, releases, and community discussion are all visible on GitHub and Discord.
Jan is free and open source, with 7.2M+ downloads reported on the site. Prices are subject to change.
Features, pricing and availability may change. Always verify details on the official website. Maintained by steven · Last updated Oct 16, 2026 →
Together AI is a full-stack cloud platform for running, fine-tuning, and deploying open-source AI models, aimed at engineering teams that need production inference without managing their own hardware.
General-purpose agent that browses and executes tasks end to end
Google's multimodal AI assistant for research, writing, coding, and agentic tasks across Google apps
LlamaIndex builds document parsing and extraction software plus open-source tools for AI applications, helping teams turn enterprise documents into usable data for AI agents and workflows.