Roboflow

Roboflow Explains Why It Sells Edge AI Hardware
Roboflow says its edge AI appliances address the operating system, driver and recovery work that companies face when deploying computer vision on hardware they manage…

Roboflow Adds Video Mirroring from S3, GCS and Azure
Roboflow has added video mirroring from AWS S3, S3-compatible storage, Google Cloud Storage and Azure Blob Storage to its Datasources feature. Teams can connect cloud…

Roboflow Brings Vision AI Reports to Team Inboxes
Roboflow has introduced Summary Reports in Vision Events, a feature that sends recurring updates from vision systems to selected recipients by email. Teams can describe…

Roboflow Adds Action Recognition for Video-Based Workflow Analysis
Roboflow has added support for action recognition, enabling teams to train and deploy vision models that identify actions across video and return a timeline with start…

Roboflow test shows how a 90-second GPU traffic spike causes failure
A 90-second burst of computer vision traffic pushed a single self-hosted NVIDIA L40S server into timeouts, while Roboflow Serverless continued returning inference…

Roboflow adds full anomaly detection support, trained on good images only
Roboflow has announced full support for anomaly detection models, which learn from images of normal items and flag anything that looks different. The models can be…

Roboflow joins NVIDIA Cosmos Coalition, adds Cosmos 3 post-training
Roboflow has joined the NVIDIA Cosmos Coalition, a group built around open world models for physical AI. As a further step, NVIDIA Cosmos 3 post-training is now…

MountAIn's Solar-Powered IBEX Camera Can Now Run Roboflow-Trained Models
MountAIn says its MountAIn Core software now connects Roboflow's community of more than one million developers to low-power microcontrollers, starting with the…

Roboflow Lets Users Fine-Tune NVIDIA Cosmos 3 Edge on Their Own Video
Roboflow now lets users fine-tune NVIDIA Cosmos 3 Edge on their own video data. The new Action Recognition project type returns a timeline of actions for any new clip…

Roboflow Plugin Lets ChatGPT Search Computer Vision Datasets for You
Roboflow has published a tutorial showing how its plugin for ChatGPT can search Roboflow Universe, compare candidate datasets and link to annotated sample images. The…

Roboflow Agent builds and deploys computer vision apps from a description
Roboflow has announced Roboflow Agent, an assistant built specifically for computer vision that recommends models, designs workflows and deploys applications that see…

Roboflow adds Google's Gemini 4 Argon for labeling and vision workflows
Roboflow has added Gemini 4 Argon, Google's newest multimodal model, to its computer vision platform. According to Roboflow, the model currently ranks first on Roboflow…

Ambarella pitches Physical AI at Santa Clara summit with X7 accelerator
Ambarella used the AI Infrastructure Summit in Santa Clara to present its Physical AI strategy, including the X7, which the company describes as its first standalone AI…

Roboflow Launches Labs, a Research Division for Real World Vision AI
Roboflow has introduced Roboflow Labs, a new applied research division meant to advance real world AI capabilities and help models see, reason and act in three…