1
Hi everyone, we have some important news to announce:
As of today, we’re no longer supporting Batch execution.
Why are we doing this? Broadly, we’re focusing our development efforts on capabilities that not only deliver strong value to our users but also support a more sustainable, commercially viable path for KNIME as a company. As part of this shift, we’re discontinuing this add-on and prioritizing development on areas that have the biggest impact for our users and customers.
Since launching cloud-native workflow automation with KNIME Pro, we’ve seen much lower demand for Batch. Automating workflows on KNIME Pro is both more scalable and more accessible to less technical users – if you rely on Batch today, we recommend switching by signing up here, where a 14-day free trial is available.
We appreciate your understanding and continued support as we focus on building the next generation of KNIME capabilities.
Hi Scott,
thanks for sharing this - obviously this can be quite a big thing for certain people.
One obvious question:
Does this mean:
With the next release, Batch Execution extension will not be available anymore
or
It will also not be available anymore for older versions?
Hi Scott,
Thanks for the info…
I share Martin’s concern on this.
I have a workflow running in a continuous loop in batch (the cryptocurrency order agent).
If I were to run on Pro or Team plan, it would eat up a lot of execution credits…
In my case I can live without it, but I understand it may be an issue for other people…
better start downloading old install archives. Every recent version which had batch already stripped from core and moved to an extension, is already affected. The Batch update side is already taken down.
too bad Knime doesnt focus on making the existing capabilities better integrated
ScottF 5
Hi @MartinDDDD -
We’re not removing Batch from existing AP installations. However, for all versions >= 5.9 there won’t be any option to download and install the Batch functionality any longer.
Will there be an alternative to installing extensions/plugins into Knime-AP via batch commands? We are continously updating a custom-written extension for an AP-installation on a remote machine via a batch command similar to the one in the attached screenshot. How can this be done with version >=5.9?
wiswedel 7
That continues to work and is not what we refer to as batch execution.
HaveF 8
It’s understandable that the company needs to be profitable. However, I think the batch functionality is still quite important. Although individual users won’t dictate the company’s development direction, I still want to express my personal opinion:
Why not consider a middle ground, such as a paid batch extension? Wouldn’t that be better?
Is it a must-have or no-go, an either/or choice? Pro might be great, but the cloud isn’t suitable for everyone. Sometimes, customers just want to run workflows locally via command line.
Software development costs have decreased significantly in the AI era. Forcing users with similar needs to the cloud simply to remove this functionality is, in my opinion, not a good idea.
Just to get this straight: if I have batch execution functioning on an existing installation, will it keep working if I perform a ‘check for updates’?
I think the answer is yes, because I am running 5.11 on one platform and there it still works. But I just want to make absolutely sure…
Thanks beforehand
Hello,
I had to take a moment to process this news. Since January 2026, I have been overseeing a research project at the University of Bonn. The goal is to transfer data from the weather stations we manage into a central database, prepare it for statistical and geostatistical analysis, and, to the extent possible, implement these analyses myself.
Since I have gained quite a bit of experience with KNIME, the plan was to use KNIME for the project and also present it on campus.
Data collection is done via REST APIs. Several ETL processes have already been completed for this. See figure.
A server environment including a PostgreSQL server is available.
The next step is to get these ETL processes running on the server. Daily and hourly execution.
An existing forecasting model should first be implemented locally and then on the server side.
Information about the project can be found here: bioIP || Home
This should clarify what types of analyses should be considered in order to provide forecasts and other results to farmers.
It’s frustrating, since quite a bit of time has already been invested and now a Plan B has to be created. Which, in turn, costs unnecessary time.
KNIME as a product and, above all, the community have always been exemplary to me. I’ve already posted quite a few times via my private account and have always received quick answers and solutions.
Cloud applications may not be used.
Or is there an alternative? I can’t seem to figure out the new models. Self-hosted? Is that something like a server application?
Best regards,
Patrik
Translated with DeepL.com (free version)
takbb 11
Hi @ScottF ,
If I’m correctly understanding this announcement, it’s quite disconcerting.
Does this mean that if users wish to stay updated with the latest versions of KNIME that it will no longer be possible to execute a KNIME workflow from the command line on the pc?
Add94 12
What about using an older version of KNIME (prior to 5.9) ? + Not sure if nodepit batch is still / will be compatible with newer versions - this could be an alternative route.
NodePit Batch — NodePit NodePit Batch — NodePit
I think up until 5.8x you are fine, starting with 5.9 it is now gone. Unless you previously had it installed as extension…
but yes for any future version it will not be available is what I took away for now.
Indeed, a sudden change in policy discovered just now :
I work in a Belgian public service and we were starting to use batch to automate some workflows. This solution is obviously far from being as optimal as KNIME Business Hub.
I’m afraid that this rather sudden decision is the beginning of a long series that goes in the direction of a shift from Knime to a proprietary tool.
The community needs to be able to anticipate this type of change better. This decision is far from trivial. It reminds me of the year around 2020, with the first decisions of Talend abandoning its community solution (Studio) in favor of its proprietary tools… Before its complete abandonment in January 2024. This led to about 2/3 of its users leaving for other solutions (like Knime in our case).
It is an important strategic choice that Knime no longer offers batch tools to automate workflows and only refers to a proprietary solution. Personally, I need to see a bit more clearly about the next steps… And in my opinion, the partners and contributors of extensions too.
We keep an eye on other open source developments: Apache Hop… or even Airflow with python nodes.
The strategic choice to switch to proprietary tools is commendable. Of course, it also ensures financial sustainability. But for me, the contributions in terms of user-friendliness and ease of use (no-code) of KNIME Business Hub amply justify the price and its overall appeal.
But from the point of view of the open source community, we need to be reassured about KNIME’s strategy in the medium term.
Hello Patrik,
Reading your post, it seems your setup is somewhat similar to mine : Marc Goossens - ClimateChange
I am using batch execution also for the ETL jobs.
I have currently updated to KNIME version 5.11, I had batch execution installed and it did not get ‘broken’ by the updates. That’s good news. Ideally would be some information from KNIME in the sense of ‘as long as you don’t do this, batch execution will keep working’ , at least for the foreseeable future. Nothing lasts forever, especially in IT.
I agree with you Nerve. The same happened to Pentaho and it’s Data Integration ETL…Actually Matt Casters, who was more or less the ‘father’ of PDI, is now advocating Apache Hop
@Add94 not sure if this will get me struck from the KNIME Christmas card list 
but I ran a short try with the NodePit Batch extension and it seems to work in a similar way like the traditional batch processes.
The current script is running on Windows - macOS is also possible.
18
Awesome, great to know! I saw their changelog and =<5.9 should be compatible, but I am not sure how this will play out in further versions. Also, nodepit batch was always my favorite way of executing workflows as scripts (supports workflows stored remotely , easier syntax etc)
Sorry I meant Herve, not Nerve… Bloody spelling corrector 