2026-03
March 30, 2026
April 13, 2026
Big update this month — we've made batch processing faster, given you more control over service versions and API call history, and shipped a handful of improvements across Spark Shell and the Coherent Assistant.
We're also retiring a few older features as better alternatives take their place, and we're excited to share a new strategic partnership with Indico Data. Details on all of it below.
What's New in Coherent
📑 Delete Service Versions
To facilitate management of Spark service versions, we’ve now included the ability to delete service versions through the user interface. This can help to clean up messy services while retaining the functionality of the "keeper" versions.
We are also working on the ability to edit service versions! Look out for this in a future release!
🔑 Deployment request access management
Spark provides an interface for users to submit a Deployment request to their internal systems from the Options page. By default, access is granted to all users who can login to Spark.
We have added a new configuration in the Tenant configuration section to manage which users are able to submit this request to improve controls around this process.
🚀 Batch execution unleashed
Spark's Batch APIs enable widescale processing of Spark services on large datasets. We have enhanced our implementation to significantly increase the default buffers that are used to manage the inflow and outflow of data. For large jobs, this makes it much easier to manage the flow of data in and out of Spark leading to faster completion times.
Learn more on how to Run High-Volume Models with Coherent's Batch APIs.
🔍 API Call History search and download updates
There may be many instances when Spark is called repeatedly for one quotation due to evaluating different options and benefits. When downloading the API Call History, a lot of extra data may be included when only the latest quotation is relevant.
We have included a new filter for the API Call History which will keep the latest combination of
source_system,call_purpose,correlation_id. This reduces the amount of data to be transferred and processed by the end user performing the analysis!In addition to this for downloading a set of API calls to Excel, we have introduced a new Fast beta option which should offer improved download speeds.nd work
🗃️ Spark Shell browser
Spark SHELL is a feature in the Coherent Assistant that introduces control and management over Excel documents.
In this update, we have incorporated a Shell browser to view all of the created Shell documents. This greatly simplifies the process of distributing the latest Shell documents to users. Instead of using network storage, SharePoint, or even emails, simply click the Shell tile in the Coherent Assistant to see all the Shells available. Upon selecting a Shell, the Excel file will be downloaded and ready to be used.
Read more about When Your Excel Logic is Ready, But You Don't Have a Front-End in our blog post on Spark Shell benefits.
Notable Enhancements
Our Analyze service with AI can work across multiple Spark service versions to enable comparison or change analysis. Open the Analyze Service functionality for any
2Spark services (or versions) and start chatting!Updates to Upload service journey to reduce the time taken to publish a service.
Improve the speed of uploads by skipping the step used to insert a cover sheet into the uploaded Excel document. This can be managed in the Tenant configuration.
Reconfiguration of service and version properties.
Coherent Assistant has additional fixes for logout issues encountered by some users.
Update to the JSONtransforms to support longer request times.
Fix for
Xreportformatting issues.Update to Neuron Excel-to-code compiler for better formula compatibility. If you use Data Tables that include dynamic headers, we recommend updating to the latest version of Neuron.
Security patch updates to addresses vulnerabilities, ensuring enhanced protection and stability of our systems.
Version number
v8.50.1
Last updated