An open-source end-to-end GenAI platform for air-gapped, on-premises or cloud VPC deployments. Users can Query and summarize documents or just chat with local private GPT LLMs using h2oGPT, an Apache V2 open-source project. And the commercially available Enterprise h2oGPTe provides information retrieval on internal data, privately hosts LLMs, and secures data.
N/A
Microsoft 365 Copilot
Score 8.8 out of 10
N/A
For enterprises, Microsoft 365 Copilot (or just Microsoft Copilot) is a generative AI operating as an intelligent virtual assistant for work. Through a chat interface, business users can use it to solve a variety of complex tasks.
$31.50
per month per user
Pricing
H2O.ai
Microsoft 365 Copilot
Editions & Modules
No answers on this topic
Microsoft Copilot
$31.50
per month per user
Offerings
Pricing Offerings
H2O.ai
Microsoft 365 Copilot
Free Trial
No
No
Free/Freemium Version
Yes
Yes
Premium Consulting/Integration Services
No
No
Entry-level Setup Fee
No setup fee
No setup fee
Additional Details
—
Pricing shown is based on an annual commitment. Discount available for annual payment.
Use H2O.ai whenever you need easy to use tool, when you must be cost efficient (you can not charge the client extra money for software licenses used), need a tool with lots of algorithms that are normally used in data analytics, or need to work on one machine (it is either not allowed to move data to cloud storage or simply not necessary to connect to Hadoop, etc.). Also, you can call H2O directly from Python which makes analysis more efficient.
In my experience, if you're using within the Microsoft office suite, it has the best integration. The usability is great and the user has to put little effort to get a task done. On the other hand, in my experience, coding within Visual Code is unreliable and the results are not consistent. You can't use with different programming languages and ask for complex tasks. Pitty because I think the VS integration is great.
This is not really a drawback, but rather a warning - the Drivereless AI is not a replacement for a data scientist yet, and will not replace data scientists in the next decade neither. The Driverless AI feature delivers reliable results only if the analyst is sure about the meaning of input data. The data quality is usually a major issue and no tool can detect the meaning of data in the input. Data scientists are also required for business interpretation of the findings. So be careful, and do not rely on this feature without a good understanding of what it really does in each step.
Like all other AI systems, Copilot suffers from hallucinations. You have to be very careful with the output it generates. It could be completely wrong. It needs to be checked and rechecked to see if it is correct.
Copilot sometimes struggles to formulate accurate responses to complex queries. It will either provide an incomplete response or generate a response that would be completely wrong.
Right now, I am unable to customize Copilot for my specific needs. Hopefully, in future versions of the AI, this will be taken care of.
It's integrated well across the Microsoft 365 suite of applications, without getting in the way, and providing useful tools. Some of the integrations need development, e.g. the Outlook add on is quite basic, and the excel and powerpoint ones can't do much yet. The implementation in Teams is really fantastic, significantly improving the experience of recording, transcription, and summarising meetings and action points.
I have used Knime, RapidMiner, and Weka before I heard about H2O, but amongst all I really liked H2O. However, nowadays Googles AutoML and AWS SageMaker AutoML platform are really competitive, but more costly than H2O.
We needed a tool where secure data was not submitted to servers to process as we have some health care related data to keep private. We also needed to be able to test this tool without committing first to see if it is a fit and Microsoft provided this easily to us.
Positive impact: saving in infrastructure expenses - compared to other bulky tools this costs a fraction
Positive impact: ability to get quick fixes from H2O when problems arise - compared to waiting for several months/years for new releases from other vendors
Positive impact: Access to H2O core team and able to get features that are needed for our business quickly added to the core H2O product
Service desk employees resolve some tickets much faster thanks to Copilot support. This percentage, according to current knowledge, is between 20% and 30%.
For administrators, creating scripts and automations with Copilot support saves them a significant amount of time. This currently stands at between 20% and 40%.
If you don't give the copilot proper instructions, you'll also get answers that aren't valuable. You do need some knowledge or training to get the right answers from the copilot.
