Build & use computer vision models fast ✨ Get started: https://nitter.cf/t.co/QvAsWtf7c8 Conference Oct 22 in SF: https://nitter.cf/t.co/YX4via3TiV
Joined June 2019
- Tweets2.3K
- Following1
- Followers15.4K
- Likes4.6K
Roboflow retweeted
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 👇
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!
if you can't make the stream, we have some of this captured here roboflow.com/blog/gemini-4-a…
Roboflow retweeted
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: youtube.com/live/bBX_c5AFuMQ
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
join the livestream tomorrow to see live demos and ask questions
youtube.com/live/bBX_c5AFuMQ
we will show you how argon can do zero-shot segmentation :)
Roboflow retweeted
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
Roboflow retweeted
Muse Wonder is pretty good vision model
suuuuuuuper expensive
vs Muse Spark 1.3
- better text extraction
- better object detection
- 2x faster
- 4x more expensive (almost Astra price)
↓ benchmarks and examples
Roboflow retweeted
My favourite thing about @roboflow Trackers is how easy we made it to swap things and create the best tracker for your case.
Swap the tracker, the IoU variant, camera motion compensation, dynamic frame rate or masks, one line change, and see what works best!
Any detector works.
At COLM? Join Roboflow Labs tonight to talk vision, language models, and everything in between.
Meet fellow researchers and builders over light bites, 7–10pm PT.
Space is limited. Request a spot:
partiful.com/e/ldk3A6Ur2of82…
Roboflow retweeted
Haiku 5.5 first look
I tested new model from Athropic on object detection and our new text benchmark
Haiku 5.5 is better and cheaper than GPT-6 Luna
leaderboards and examples ↓
playground.roboflow.com/eval…
Roboflow retweeted
openai just released Decisions API, Jev-like "system 1" multi-modal model based on GPT6 Luna!
Each egg gets cropped and sent to OpenAI (~270ms latency) and returns P(clean, dirty, cracked).
One image crop consumes ~280 input tokens (~$0.03/1K imgs). @roboflow RF-DETR for detecting eggs
Roboflow retweeted
I spend a lot of time finding out what VLMs still can't do
pretty excited to help change that
we launched Roboflow Labs at @roboflow
training data, evals and environments for real-world tasks. vision, multimodal reasoning, robotics and coding.
link: blog.roboflow.com/roboflow-l…
Roboflow retweeted
I just labelled over 1000 segmentation masks for utility poles in under 90 s 🤯
-GPT Astra for box detections
-SAM3 for masks on those boxes
How long would it take a human to do the same??
GPT-6.1 Sol is going to give you similar accuracy at 20% of the cost if you are ok with 3x slower response time
Some really can things can be built with this level of intelligence if the latency works for your use case
Pumped to see what people build with it
GPT-6.1 Sol is almost as good as Astra on object detection (and segmentation)
- #2 on our detection benchmark
- 2% behind GPT-6 Astra
- 78% lower cost
↓ leaderboards and examples
GPT 6.1 Sol for video analytics, watching a full airplane turnaround ✈️
- Sol is near-Astra level at vision at 1/5th the cost, it matched every event Astra found on
- GPT 6.1 Luna/Terra were >30% less accurate on action recognition
- Sol also detects ground vehicles (SAM3 on @roboflow tracks them) and logs all states (jet bridge attached, catering at the door, tug connected)
- Bags on vs off and deboarding vs boarding come from optical flow at native 30fps (cost optimization)
- Whole 40min turnaround for ~$1