Bluespeed AI

[castrojo]

"Bluespeed"

Mission: Drive open source AI adoption by delivering a kickass community driven desktop experience.

Initial cut: ujust bluefin-ai top level just like bluefin-cli to turn all of this on.

• Include goose cli
• Include goose desktop?
• Light shell integration (needs details)
• Pulls small LLMs for on-device work
• "Some sort of goose config" that has this work ootb. It should also be trivial for me to connect anything goose supports.

Awesome things we want that should be scoped out seperately:

• "Ask Bluefin" chatbot that talks to local llm. This should be exposed via a ramalama openai endpoint so that the user can swap in an LLM of their choice
• dosu cli or some other desktop integration
• Figure out how to ship/use the linux system MCP RH is working on: https://github.com/rhel-lightspeed/linux-mcp-server. Anything agentic should be a "power mode" since it changes the safety model. Readonly for diagnostics and introspection is a clear boundary that we should enforce.

Users are using general LLMs for Bluefin and those won't be as nice as a custom solution. With Ask Bluefin you can use it for everything. For example I don't think I've written a service unit by hand in like 6 months, everyone should have this.

Ideas

Hosted Ones

They don’t need to be perfect, they just need to be better than the average redditor. This is mostly done. 🙂

These are fantastic tools but we also need to:

• Ensure local first OSS is the halo experience, remove third party dependencies
• Give our experts a platform to experiment
• Align with Red Hat’s AI initiatives to share resources
• Do for RHEL Lightspeed what Universal Blue does for bootc
• Build a community
• We can curate some awesome OSS stuff since we can curate our user experience
• Brings more open source AI people into this, we’re selling the idea so we need everybody.
• We can help GNOME by proposing that desktops should have API standards around AI so developers can build applications

Bluespeed

Local AI controlled by the user, (spiel). We build on ramalama and have the capability to ship custom OCI images from ghcr already, infrastructure is in place! Ideally the tools install cleanly, no background services unless enabled, endpoint is local-only by default, and the uninstall path leaves no surprises.

• We already ship ramalama
• We declare that our images will have standardized API endpoints driven by ramalama, opt in due to tech (llm size, etc)
• We pick a base model to ship, small because laptops - We RAG in official docs, source code of everything, Homebrew docs, Flathub, Bazaar, ramalama, podman, podman desktop, bootc, devcontainers, vscode, etc. You get the idea. The RAG should probably be a seperate deliverable, bluefin-knowledge or something.
• Final goal of proposing AI endpoints to the major desktops for standardization.
• We maintain the prompt in github to be community maintained:

This also means the pattern can be templated. Imagine the universal blue custom image template having this ootb - an organization’s custom image could ingest everything and they have the structure in place to do what they want. Bazzite could ingest sites like: https://steamdb.info/ and https://www.protondb.com/ so that users can have fresh information on things -- these are crucial resources for new users!

Implementation

We add “Ask Bluefin…” to the menu. We use alpaca or another GTK chat app thing, user hits enter to accept the thing so that it downloads the right LLM stuff, managed by ramalama in the backend. We bind this to a copilot key if it exists, and bind it to F1 or some other help shortcut. We include a PDF of the docs on the installation anyway. And yelp was removed from GNOME, this is the replacement.

Shell and Other Shell integrations

We should strive for “enthusiast driven version of RHEL Lightspeed”. Light shell completion perhaps? Open to ideas. Should be very light, don’t push it, they need to be softly onramped. They are probably anxious.

Pros should be able to “opt-in” to powerful mode that can prototype ideas and go full hog. Ideas proven in Bluefin could be useful to use somewhere else. We need to find and ship the best VSCode extension out of the box on our developer images. Something like continue? But then the pile of 3rd party deps piles up. Needs discussion

Why?

We will earn a great reputation with open source AI experts, this is a huge growth area. Those hills are rich in new open source contributors waiting for the right messaging:

• We will stick to enabling APIs with Ramalama and not building wiz bang products, we offer an AI API for app developers - we want more alpacas and Jans.
• We’ll ship enough oss goodies so that other OSS vendors can get wins as well, we’re trying to lift all boats here.
• Distros gonna distro, none of them are going to touch this stuff because they’re going after 1990s linux guy as their user base and they have no clue how quickly open source AI has already surpassed the free desktop world. Sigh.
  • We don’t even call it AI, we just talk about the features, aka “Ask Bluefin”, there’s nothing for them to argue about.
  • We repel the usual suspects
• Ramalama hardware accelerates a ton of stuff ootb, it’s a breeze to get it up and running vs. ollama.
  • Since it’s containers all the cuda/rocm containers get updated via the existing update tools transparently to the user anyway
  • All those security benefits of doing it the container way.
  • What's the tldr on this at RH now that Dan is retired? Is there a champion? Find them.

Not Interested In

• AI bro tech, we should scheme with people who are OSS-first.
• Arguing with 1990s Linux guy.
• Shipping image gen AI things. The artists paid for Bluefin’s artwork have had their lives affected by AI, it’s the only reason I can afford them. I want to make an implied statement that we provide API endpoints and drive open source, and yet Bluefin is a human creation -- how you view this is your decision to make, not ours.

[joshyorko]

@castrojo So you’re proposing a Cloud-Native Blueprint for local AI.
​The killer app here isn't just the feature; it's the Template.
​Every Bazzite user could have a local AI that knows exactly which GE-Proton version runs Elden Ring best.
​A Corporate team forks it, ingests their private Wiki, and ships a versioned "Ask our Docs" agent that never touches the cloud.
​Obviously, it’s so much more, but I'm diving in to build the prototype. Looking at the Red Hat 2026 Agentic Roadmap now

[castrojo]

We're just keeping it simple and shipping things that rhel lightspeed folks are working on (but can expand from there!). We wouldn't brand any of this it'd just be ujust troubleshoot on our systems or something like that.

Other people will write the apps (GUI/cli) for this eventually anyway, our primary purpose is to get it out there in a way that we can start playing with it everyday.

[sheikhlimon]

Yes, the shell variable expansion issue is fixed upstream. If anything related still comes up during integration, feel free to open an issue and we’ll be happy to take a look.

[castrojo]

@joshyorko Please don't paste in all that stuff from your LLM, it's too noisy and is just repeating what we already know, thanks!

@sheikhlimon thanks for dropping by! (I've joined your discord and we're going to take a look at helping with the flatpak for linux!).

It's been a late night since we got it working and started playing with it. Basically we've already got an MVP with a goose front end preconfigured to use with linux-mcp-server. We can totally ship this now in a ujust and it'd be a pretty dope first effort.

We decided to write the docs of what this would look like ahead of time on a hidden page:

https://docs.projectbluefin.io/troubleshooting/