Self-hosted AI had a big week: a private workspace, a memory layer, and a local ChatGPT clone
Four tools caught my eye this week, and they all point the same direction: stop renting your AI assistant from a cloud vendor and run it on your own hardware. The privacy pitch is finally backed by tooling that doesn't take a PhD to set up. odysseus is the one everyone's talking about. It's a self-hosted AI workspace that bundles chat, autonomous agents, email triage, and research into one app, and it can serve open models directly. The clever part is the Cookbook: it looks at your actual hardware and recommends models you can really run, then wires them up for you. It launched to a massive audience overnight, so expect rough edges. And read the security note first. The project itself tells you to treat it like an admin console and never put it on a public IP. The other three fill in the stack. supermemory gives your agent a long-term memory that survives across conversations. Open WebUI is a polished ChatGPT-style interface for whatever models you're running locally. DeepSeek-GUI puts a real review step in front of an AI coding agent, so you see the diffs before anything changes. The common thread: you own your data, and you pay for it in hardware and setup time instead of a monthly bill. Whether that trade is worth it comes down to what you've got sitting under your desk.
Self-hosted AI workspace.
The Lens
Odysseus is a self-hosted AI workspace: chat with local or hosted models, autonomous agents with tools and MCP, deep research, blind side-by-side model comparison, a writing editor, an email inbox with triage and reply drafts, notes, tasks and calendar sync. Its Cookbook looks at your hardware, recommends models that fit, then downloads and serves them for you through vLLM, SGLang, llama.cpp, MLX or Ollama. It's AGPL-3.0. Docker Compose gets it running at localhost:7000, with NVIDIA and AMD GPU overlays; pin the immutable image tag rather than latest. On Apple Silicon, Docker can't reach the GPU, so use the native install. Its own threat model says to treat it like an admin console: a logged-in admin can run shell commands, read and write files and send email. Keep auth on and never expose it directly to the internet. No paid tier; your costs are hardware and any hosted-API tokens. Solo tinkerers with a decent GPU get a lot for free. Small trusted teams on a private network can share it. Anything bigger needs a security review first. AGPL means that if you modify it and let others use it over a network, you owe them your source. The catch: it's young, the issue tracker is enormous, and it touches your email and your files. Without a capable GPU or plenty of RAM you're limited to small local models.
Memory engine and app that is extremely fast, scalable. The Memory API for the AI era.
The Lens
Supermemory is a memory and context engine for AI apps. It extracts facts from conversations and documents, keeps a profile of each user, and runs hybrid search that blends RAG with personalized recall, all behind one API. The company reports top scores on the major memory benchmarks. The main repository is MIT, and plugins exist for Claude Code, Cursor and Codex.
Running it locally is one command, npx supermemory local, which starts a single binary with the same API as the hosted platform. It works fully offline with a local model through Ollama, and your data stays on your machine. The local server is single-user and licensed for up to 10,000 documents, though, and the connectors for Google Drive, Gmail, Notion and OneDrive, plus the Supermemory MCP, only exist on the hosted platform.
Hosted pricing is published and credit-based. Free includes $5 of credits a month and pauses when they run out. Pro is $19/mo with pay-as-you-go top-ups, Max is $100/mo, and Scale at $399/mo adds S3 and web crawler connectors, unlimited seats and a self-hosted option. Teams building production memory will land on Pro or Scale.
The catch is maturity. The local server is still version 0.0.x, and one August release was an emergency patch after an upgrade silently wiped search vectors. The consumer app and browser extension were retired in September too. Build on the API, pin versions, and back up your data.
AI agent workspace for DeepSeek models, with Code and Claw modes built into your application.
The Lens
DeepSeek-GUI wraps DeepSeek's AI coding agent in a desktop workspace instead of a terminal. You point it at a real project and it edits files, shows you the diffs, and waits for your approval before changing anything, with permission levels from read-only to full access. MIT-licensed and free to install. It's for people who want DeepSeek working on actual code with a review step, not just a chat window. It ships three modes. Code mode is the project workspace with file monitoring and task cards for jobs like debugging. Write mode is a Markdown editor with inline AI and export to HTML, PDF, or DOC. Claw mode runs background automation, hooking into Lark/Feishu, webhooks, and scheduled tasks. It also manages MCP servers and skills. Prebuilt installers cover macOS and Windows; Linux builds from source and needs Node 20 or newer. The app is free, but it's a front end, not a model. You bring your own DeepSeek API key, and that usage costs money, though DeepSeek's pricing sits among the lowest of the major providers. Solo developers already using DeepSeek who want a GUI with approval controls: worth a look. Teams get visibility into what the agent changed before it lands. The catch is that this is one developer's project tied to one model provider. You're betting on both the maintainer keeping pace and DeepSeek's API staying available and cheap. If you want a model-agnostic agent workspace, look at cmux or the broader AI coding tools that aren't locked to a single provider. Outside the DeepSeek ecosystem, this does nothing for you.
Self-hosted AI interface for LLMs
The Lens
Open WebUI is a ChatGPT-style web app you run on your own server. Point it at Ollama, LM Studio, vLLM or any OpenAI-compatible API, and your team gets chat, document Q&A (RAG: the AI answers from your files), web search, voice calls and per-group permissions. It's free to run.
Setup is pip install open-webui or one docker run. The models behind it are the real hardware or API bill. LDAP, SSO and SCIM ship free. Releases land every week or two, so pin versions.
The paywall is mostly about the logo. Keep the Open WebUI branding and use is free at any size. Past 50 end users in a rolling 30 days, white-labeling needs an enterprise license, and so does Terminals, the per-user sandbox add-on. Prices aren't published. Solo and small teams: free. Larger teams: free unless you rebrand or want Terminals and an SLA.
The catch: security. The project has published more than 150 advisories since February, many rated high: account takeover, session-token theft, SSRF. Most need a logged-in user, so patch fast and keep it off the open internet. Also, the license isn't OSI open source; LibreChat is the MIT-licensed comparison.
Get the next issue in your inbox
Free. No spam. Unsubscribe anytime.