assembled by @FactoryAI
curl -L app.factory.ai/cli |sh
Joined December 2025
- Tweets684
- Following1
- Followers8K
- Likes3.6K
Pinned Tweet
Make software that improves itself.
👉 factory.com
Droid retweeted
If you’re not sure whether you’re getting the most out of your coding agent spend, shoot me a note or comment in thread and we’ll show your team what it should look like with a 1 mo free trial of Factory.
Best performance, best cost efficiency. Choose both.
Droid, but better!
Droid iterates until performance improves. 🫡
More massive performance gains thanks to @droid working tirelessly. I need to get a little plush of my droid friend
Nothing beats hearing this from a long-time @FactoryAI user. 🫡
can you share this benchmark methodology?
i've been using @droid as my harness since mid 2025, try everything out there, opencode, omp, amp, etc etc, but nothing else has being better on my experiences
even tho, I'm always open to new tools, and you get me in this Command Code showcase with v4.1 flash thats currently my workhorse alongside opus 5.5.
i would like to include droid in this comparisson versus command code harness
How Blackstone is building their own software factory with @FactoryAI: Read the full story below 👇
If you had to guess, how agent-ready is your codebase?
Well, you don't need to guess anymore. Test your Agent Readiness with @FactoryAI.
The entire game was built code-only with @FactoryAI. 🎮
Check out the original post for the live demo link & prompt:
What do you plan to do after delegating all your work to @FactoryAI?
See what Delegations can already do for you: nitter.cf/FactoryAI/status/21111…
Any Droid builders in London? 🇬🇧
I was this last days at London, and suddenly at Piccadilly I saw @droid @FactoryAI Wow!
Congratulations!
Droid retweeted
Did a sanity test in @apolloaiapp and planning to serve this via with @droid harness.
Run DeepSeek V4.1 Flash 1017 on 2x DGX Sparks ✨
- 1M total KV pool, 512K context
- 75 tok/s on prose, single stream
- 130 tok/s on prose, 4 concurrent streams
- Prefill 2000-2175 tok/s
- EXL3-EAQv1, 2.9 bpw
The quant in this recipe got about the same calibration as my GLM 5.3 Flash, with some extras. Feedback would be appreciated!
Special thanks to @bertholomusai and all the contributors for the engine work, and to @ashxhart for TensorFold.
Get it here:
github.com/MiaAI-Lab/DeepSee…
Droid retweeted
Built in @droid with Argon!
Droid retweeted
Polishing that @Google Map BillBoard graphic display that @FactoryAI’s @samwilhoit_ created for the epic Billboard demo 🔥
@bot driving screen caps
Droid doing @droid things
to understand the placement of the panels in 3d space for the more complex multi-panel layouts, I created a custom 3d visualizer tool to map the billboard video panels onto 3d buildings in google maps. this was a game changer compared to trying to digest 2d spec sheets for each location and adapt each panel design.
I fed the billboard spec sheets containing the exact panel sizes, placement, locations etc into @FactoryAI and enabled Google Maps 3d tiles API, and the coding agents were able to place the artwork onto the buildings in the exact configuration and adapt each creative to spec.
What is your model mix? Do you optimize for cost, speed, performance, or something else?
Day 16 on the @droid Max plan ($200)
Wanted to know if GLM 5.3 Flash and Nemotron 3.5 Super are actually worth using in Droid, since they're both super cheap in the picker (0.06x and 0.08x). So I benchmarked them.
23 Python tasks from @VulcanBench, 3 runs each, and hidden tests decide if a task counts as solved. Ran DeepSeek V4.1 Flash and GPT-6.1 Sol next to them so I had something to compare against. 690 runs total and my laptop lowkey almost died doing it 😅
Sol is on top at 87%, but it costs 26k to 46k credits a task.
The surprise was GLM 5.3 Flash on low. 80.6% solved at about 5.7k credits a task. That's half of what DeepSeek low costs and about a fifth of Sol low. Easily the best value in Droid right now.
DeepSeek V4.1 Flash on high is still the sweet spot if you care about getting it right. 86.8% for about 14k credits, so almost Sol level for around half the price, and it's fast too.
Nemotron 3.5 Super was the disappointing one. Same score as GLM on high but almost 3x the credits. Honestly couldn't find a job where it wins.
So this is how I'm splitting it now. GLM 5.3 Flash low for exploration, mechanical edits and anything high volume. DeepSeek V4.1 Flash high as my default coding worker. Sol only for the hard stuff where getting it right matters more than credits. GLM max barely helped (82.3% for almost 3x the credits of low) so I'm skipping it.
It's one Python suite, so take it with a grain of salt.
Check out this demo from @adityavg13👇
Using Gemini 4 Argon (Medium) in @droid, I wanted to start to share demos from my program, emath, that I'm working on
#1 Structures & Worlds