Roboflow

Turn images, videos, and streams into intelligence

[Roboflow @roboflow]

Retweets are not endorsements, but this one is 🦝🦝🦝🦝🦝

[Nintendo of America @NintendoAmerica]

Live your best raccoon life in this chaotic co-op sandbox for 1-4 players! Complete crazy missions, cause chaos, and outsmart a relentless Exterminator. 🦝

Dumpster Gang is coming to Nintendo Switch 2 in March 2027. #IndieWorld @DumpsterGangHQ

[Roboflow @roboflow]

Joining the coalition places Roboflow Labs’ work within a shared effort to advance those capabilities.

We see an opportunity to connect progress in frontier models with the breadth of tasks developers face outside the lab, and to help turn that progress into tools people can build with.

Roboflow Labs draws on what the 2M+ developers and Fortune 500 enterprises building on Roboflow are actually trying to solve.

That real-world task distribution shapes the research, benchmarks like RF100-VL, and data we bring to model builders.

More on Roboflow Labs: https://t.co/v5KcLQv4xx

[Roboflow @roboflow]

Today we’re announcing that we’ve partnered with NVIDIA and joined the Cosmos Coalition, a global initiative among leading AI labs, as a premiere provider of multi-modal data and post-training capabilities that advance frontier world model research.

We believe the next wave of the AI boom will be defined by what intelligence can do in the physical economy, and that vision is the bottleneck to unblock.

[Roboflow @roboflow]

NVIDIA introduced the coalition in May 2026 to bring model builders, developers, and organizations building physical AI together around open world models.

The effort supports collaboration on models, research, and evaluation methods to advance systems that can understand and simulate the physical world.

More info: https://t.co/bgpK0LT8Wh

[SkalskiP @skalskip92]

GPT-6.1 Terra beats Astra in our vision benchmarks

wtf OpenAI?

had to double-check the results

- text
- object detection
- counting

also ~47% faster and ~92% cheaper

↓ leaderboards and examples

[Roboflow @roboflow]

We are hosting the Visual Intelligence Summit on October 22 in SF.

The physical world is AI's biggest opportunity and the best way to advance humanity in our lifetimes. Few people are building for AI beyond the screen.

The Summit is for the people who see it differently and want to be at the frontier of bringing AI into the physical world.

If you’re building for that future, join us.

[Aurick Anwar @007Aurick]

Ever wanted to know if you're cooked in an IRL game of chess? 💀

So I built Stockfish IRL. A camera watches a real board, tracks every move, and literally yells at you when you blunder.

Built with @roboflow 👇

[Roboflow @roboflow]

@MrrrOzi but you can't ask questions on the rerun!

[Roboflow @roboflow]

livestream today sharing all the findings we have on using Argon for vision

TLDR: Google is back. Argon tops almost everything.

Argon is good enough that we started testing it on new areas where most VLMs aren't usable like segmentation, keypoints, and video reasoning

Exciting to have a model this good. Never been a better time to build with vision

Also, models hitting 90% on our evals means we are going to invest in expanding benchmarks. Stay tuned!

[SkalskiP @skalskip92]

Gemini 4 Argon stream in 30 minutes. see you!

[Roboflow @roboflow]

@lillyguisnet we didn't develop the COCO benchmark so no plans to expand that baseline

RF-DETR is made to be fine-tuned on custom data and not meant to be used as general detector

using VLMs to label data then train a custom RF-DETR is the way to go!

[Roboflow @roboflow]

if you can't make the stream, we have some of this captured here https://t.co/CFzOJa9cw1

[Roboflow @roboflow]

@lillyguisnet considering coco is only 80 classes, very funny to have teddy bear make the cut!

[Roboflow @roboflow]

@ajs6888 yes, have to have evals!

[Roboflow @roboflow]

Gemini 4 Argon expands what’s possible in computer vision

As models improve, your apps should too

Routing tasks to the right frontier or specialized model is how you can build production systems that scale with your accuracy, speed, and cost needs

Can't wait to see all the new products people ship with this one

[SkalskiP @skalskip92]

Astra dethroned

Gemini 4 Argon is the best "vision" model

#1 for detection:

- tight, precise boxes
- dense scenes

#1 for text extraction:

- small text
- handwritten text
- complex tables
- rotated text

4× cheaper than Astra at high

[SkalskiP @skalskip92]

Gemini 4 Argon vs GPT-6 Astra YouTube stream

tomorrow, Oct 15th, 8 AM PT / 11 AM ET / 5 PM CET

save the time and see you tomorrow

- detection
- segmentation
- transcription
- data extraction
- efficiency and cost

what else?

link: https://t.co/Ww478jg9tb

[Roboflow @roboflow]

join the livestream tomorrow to see live demos and ask questions

https://t.co/bY3N09N9A5

we will show you how argon can do zero-shot segmentation :)

[Roboflow @roboflow]

Here's how we can help you do that https://t.co/5UfDDwi1V2

[Roboflow @roboflow]

@shelbrock_kevin @grok Try Auto-Label in Roboflow; you probably won’t have to label something like this by hand these days. (You could also probably just hook our MCP up to Grok and let it run things for you.)

[Roboflow @roboflow]

@lillyguisnet ahhhh super cool, thanks for sharing a bit more about what you're evaluating.

this is the overall score combining all evals across tasks

thanks for sharing your results! excited to see more for future models

[Roboflow @roboflow]

Muse Wonder coming in at #8 on our evals and tied for most expensive model in the top 10, Astra is #1 and same cost

[lillyguisnet @lillyguisnet]

New sota for my image pointing