
Research Scientist @LumaLabsAI. Core contributor to Uni-1. Now doing RL for robotics
Joined June 2015
- Tweets30
- Following105
- Followers118
- Likes279
Fable 5.2's launch impressed with strong robot manipulation. How does ZAP, our 27B model, do on the same tasks?
We recreated the tasks in sim and ran both through the same harness (the model is the policy, calling robot tools directly). ZAP only trained on sim pick-and-place, so every object and goal is new to it.
Task 1, open a bottle: ZAP is 2.5x faster and costs $0.43 vs. Fable's $6.50. It plans the grasp up front, and it pays off: a firm hold and clean turns. Fable drops the bottle, recovers nicely, and still finishes.
Sim only for now since I'm at CoRL, but the real robot is waiting back home 👀 🧵
1/4
Task 3, assemble a gear set: fit a gear into place so it meshes with the others, then turn the center gear. This one demands precision, and Fable shines with cleaner grips and more precise placement.
ZAP's placement is less tidy, but it's nearly 2x faster, takes fewer turns, and costs $0.24 vs. $2.54.
3/4
1
1
105
The last task in their demo vid, solving a Rubik's cube, is still on the to-do list (I need to build the sim cube first 🧩).
Scorecard: both models finished all 3 tasks. ZAP was 1.3–2.6x faster and 7–15x cheaper, but Fable was cleaner on 2 of 3.
Still plenty of work left for agentic robots on precision, cost, and speed. More soon!
zero-real.github.io
4/4
3
88
Come say hi! Presenting ZAP at 12-1:30 at Griffin Hall, poster 134
Who says robot control needs a giant model? ZAP, a small VLM, matches Astra on real robot tasks - 3× faster and 17× cheaper.
Presenting tomorrow at #CoRL2026. Come chat!
📍 Griffin Hall · Poster 134
🗓️ Mon, Nov. 9 · 12:00–1:30 pm
Who says robot control needs a giant model? ZAP, a small VLM, matches Astra on real robot tasks - 3× faster and 17× cheaper.
Presenting tomorrow at #CoRL2026. Come chat!
📍 Griffin Hall · Poster 134
🗓️ Mon, Nov. 9 · 12:00–1:30 pm
Astra made robot control look like a job for frontier LLMs.
But small VLMs already know what to do. They just don't know how. Turns out, simulation is all you need.
Meet ZAP. No real robot data, no frontier teacher. Matches Astra on real tasks, 3× faster, 17× cheaper. 🧵
(1/11)
You don't need a frontier model to control a robot. An open VLM already knows what to do; simulation can teach it how.
Learn more at zero-real.github.io
(10/11)
The awesome team behind ZAP at @LumaLabsAI: @djsaunde, @mirceamironenco, Salman Mohammadi, @doublemathew, @srzhu97, @_DivyaMakkar, @xuhuahuang01, @qinboli, @nsupreeth5, @censiyuan1, @Ir1dXD, Shyamal Buch and @thomasneff93
(11/11)
10
832
Ben ✈️ CoRL 2026 retweeted
Very excited about this! We've been cooking internally for to try and build a very fast & exact sequence parallelism strategy. TL;DR: it's really hard to find the one strategy to rule them all, but based on thorough evaluations and kernel work, we found a very strong overall approach :)
We'll be at PyTorch conference tomorrow and wednesday - if you want to learn more, please reach out! Also code & a more detailed technical report on this will follow!
teapawt.net/LumaLabsAI/status/2112…
Exact sequence parallelism for packed multimodal masks.
It picks ring attention, a sparse all-to-all or Ulysses per layer, matches single-GPU numerics, and beats a FlashMask + Ulysses hybrid in 47 of 48 cases on up to 128 H100s.
lumalabs.ai/news/packed-mult…
Ben ✈️ CoRL 2026 retweeted
Language models as forecasters can be accurate while being badly calibrated and overconfident. Their chain-of-thought can also omit the evidence that actually changed a forecast.
In this new work, we find that internal activations give us a much more reliable signal! 1/6
Can LLMs predict the next World Cup champion?
Goodfire partnered with @EternisAI to improve how LLM forecasters use available evidence and manage uncertainty.
We found models were overconfident in their predictions – but probes significantly improved calibration. (1/6)








