alsima

18 ms·

1.
by alsima 1mo ago
Thanks! Hopefully for synchronous tasks that would take CC/Codex >30 minutes we're looking for a ~30% speedup so feedback cycles are faster.
2.
by alsima 1mo ago
Hmmm couple things you can do I think OpenCode charges a subscription for models you can use in app without porting over a subscription or API key from CC/Codex, or you can charge enterprise clients, but that's something we'l
3.
by alsima 1mo ago
We've since fixed this! There's now an options in settings to remove Bullet as an author.
4.
by alsima 1mo ago
Understood, sorry to hear that! We're actively working on a guest mode for people who prefer to not make an account.
5.
by alsima 1mo ago
We don't charge any subscriptions or fees. We're not open-source, but definitely considering it, has been highly requested. Business model TBD, right now we're focusing on making a product people love.
6.
by alsima 1mo ago
Hi that makes sense, and yes, will definitely review the security permissions requested by our Linux app, sorry to hear that it was hard to trust. We're definitely thinking about open-sourcing, it's been highly requested.
7.
by alsima 1mo ago
Yes, the CLI can run in Docker. The desktop app is not designed to run in Docker. And yes, you can use any HTTPS OpenAI-compatible endpoint, including a custom base URL and model ID.
8.
by alsima 1mo ago
Really cool that you built your own harness off of mini-swe-agent, we took a lot of inspiration from that. We're looking forward to trying out Sol Ultra Fast ourselves too! In our own experience, testing, building, searching, and other
9.
by alsima 1mo ago
So sorry about that! We disabled download on iOS, but not Android (we'll fix that), it's not meant to be used on a phone (except for using our remote control feature).
10.
by alsima 1mo ago
No, we like that we're making this for ourselves as well!
11.
by alsima 1mo ago
Thanks! Yes, adding MCP support and preserving skills is definitely a priority for us right now.
12.
by alsima 1mo ago
Thanks for trying it! If you want to time it against Claude/Codex in app, we have a race lab you can use for that. Good to hear that the Linux support and integrated browser help.
13.
by alsima 1mo ago
Thank you so much!!
14.
by alsima 1mo ago
Agree, building real apps with the harness is the strongest proof. Would love to hear where Bullet fell short vs your setup, especially on speed for your builds.
15.
by alsima 1mo ago
We're actively working on implementing a guest mode feature right now for users that don't want to make an account. For reporting bugs, we have an in-app feedback form. Bugs are not automatically shared.
16.
by alsima 1mo ago
Yes, we saw this and it's really exciting, can’t wait to try it in Bullet. But faster inference only speeds up generation. In our own experience, testing, building, searching, and other tool work still dominate plenty of real tasks. We
17.
by alsima 1mo ago
No, we don’t support ACP today. Bullet uses its own internal harness and CLI integration. That said, we can definitely see it being something we work on in the future!
18.
by alsima 1mo ago
Thanks!
19.
by alsima 1mo ago
This is a great point, we're actively working on a fix. If you’re up for giving it another shot, would love to hear other feedback on the app you have.
20.
by alsima 1mo ago
Thank you!
21.
by alsima 1mo ago
Mentioned in my comment above. Sorry, I thought you were talking about Gemini 3.6 Flash, not their new one. Yes, that's the reason why we didn't include the router in the benchmark to have a fair comparison of the harness itself.
22.
by alsima 1mo ago
Fair point, yes, the router is only in the live system (we'll try to make that more clear). We thought that comparing the router to other harnesses on the benchmark wouldn't be a fair evaluation because the speed/cost gains f
23.
by alsima 1mo ago
Thanks for the feedback, we do focus on good tool management, context efficiency, etc. As for routing, ideally it isn't the user who has to route themselves, but a layer that routes for them. Thanks for the link to the repo! We have lo
24.
by alsima 1mo ago
I think that if there are many players in the space, it's a problem worth working on! Like I said in the post, our main problem is speed and that's what we're trying to fix. It's been a problem, and it still is a problem
25.
by alsima 1mo ago
Like I mentioned, we omitted routing from the benchmark and evaluated with our strongest model modality. Also, randomly selecting between Fable and Gemini 3.7 Flash wouldn't preserve quality.
26.
by alsima 1mo ago
We're optimizing every day :)
27.
by alsima 1mo ago
The benchmark wasn't evaluated with the router, we wanted an apples-to-apples comparison of our harness against other harnesses like mini-swe-agent on the same models to see if we added speed + cost value beyond routing.
28.
by alsima 1mo ago
We'll get on that :)
29.
by alsima 1mo ago
Thanks for bringing this up and trying it out! We've disabled Cmd+Option+I in the new version and also have a message during signup about sharing chats with Bullet, let us know if you run into any other issues!
30.
by alsima 1mo ago
Thanks for the detailed feedback, this was really valuable. We fixed file autocomplete, automatic CLAUDE.md loading for Claude subscription sessions, and the plan issues. Plans now only appear when requested, and progress only updates after

More ›