0.7.0 New

Compute credits on OvalBee Cloud. Node runs and deployments are charged in credits for the time they run, at the price per hour shown for each machine class. Running out of credits for the month stops the workspace’s running work and refuses new runs until the credits reset; scheduled runs and runs started by an upstream node are skipped. See Compute.
Machine classes on OvalBee Cloud. CPU (2 vCPU, 8 GB RAM) and whole GPU cards: GPU · 12 GB, GPU · 24 GB and GPU · 32 GB.
Runs at once per machine class. A plan sets how many runs of each machine class a workspace can have going at the same time, shown as max N at once next to the price. A run over the limit is refused until one of them finishes.
Credits in the sidebar. When a workspace’s credits run low or run out, the sidebar shows what is left and a banner explains it on every page.

Improvements

Labeling and Review nodes show live labeling statistics on their card.
A Labeling or Review outcome dataset without an ontology uses the ontology picked in Classes, so every outcome shares one set of classes.
A node run shows Preparing launch while its machine is being set up, and turns running when the node actually starts.
The Billing page shows the current credit burn and roughly when the month’s credits run out at that rate.

The labeling tool opens items whose tags carry a value.
Dataset insights and annotation quality reports no longer list a report that found something under “nothing found”.

Export to cloud storage. The Export to cloud storage node writes a dataset’s images, and their annotations in Supervisely, COCO or YOLO format, to an S3-compatible bucket. If files exist chooses whether to overwrite, skip or stop on files already in the bucket.
The agent runs flows. An AI agent can now start a flow run, follow its status, wait for it to finish, and read the task logs, reports and report files it produced. Before starting, it names any upstream node that has not produced a value yet.
Hardest images. The Evaluate object detection report shows the images with the lowest F1 first, each opening the ground truth beside the predictions. Sample images similar to failures starts its search from them.

Labeling, Review and Consensus review start with default Outcome statuses: Submit and Skip for Labeling, Accept and Reject for the review nodes. A status left without a dataset gets a new one on the node’s first run.
A dataset input wired from another node’s output uses the dataset as it is when the node runs, instead of the version the producing node wrote. The node’s preview and sidebar show the dataset it will run on. The dataset picker can now select the first version.
Align layout on the canvas uses a new layered layout, and parallel connections no longer draw over each other or over nodes. Nodes the agent adds are placed the same way.
The agent’s flow edits move only the part of the flow they change, and it can add notes and change connections.
Prelabel with live training previews its output dataset on the node card.
Class colors of YOLO annotations no longer change between views.
Publish… in the flow menu is disabled outside the public flows workspace and says why, instead of opening a dialog that fails.
On OvalBee Cloud, a new personal workspace starts on the Pro plan during the beta.

Train/validation split applies its default ratio and shuffling again. A node added while they were missing keeps the empty values, so add it again.
Task history opens the datasets a past run produced, not their newest version.
The plan shown after signing in as another user in the same tab is no longer the previous user’s.

The Dataset source verification node is gone, and Dataset preview no longer appears in the node picker. The Dataset node shows a dataset’s assets and annotations on its card.