Join our daily and weekly newspapers for exclusive content on the latest updates and industry-composure AI coverage. learn more
Productivity platform perception Taking bets on the big language model (LLMS). New enterprise capacityOpenai’s GPT -4.1 and Anthropic’s Cloud 3.7 include them in their dashboard.
Also as both Openi And anthropic Start building productivity features in your related chat platforms, showing these LLMs into a separate service shows how competitive the space is.
Dharna today announced its new all-in-one toolkit inside the scope, including AI meeting notes, enterprise search, research mode and the ability to switch between GPT-4.1 and cloud 3.7 of Anthropic.
One of the new features allows the user to chat with LLM inside the workplace and switch between models. Right now, the perception supports only the GPT-4.1 and Cloud 3.7. The idea of reducing window and reference switching.
The company said that the initial adoption of the new feature included Openi, Ramp, Versel and Harvey.
Model mixing and fine-tuning
Dharna produced features with a mixture of LLM from Openai and Cloud, and its own models.
The pure argument makes the model interesting to the model away from the model. GPT-4.1 is not technically an argument model, while Claude 3.7 is a hybrid model- capable of acting as a regular LLM and a logic model.
Reasoning models are in an instant, although many warn that these models can still lie sometimes. However, Openai’s O3 (and yes, Cloud 3.7 Sonnet), arguing, takes their time to respond, passing through various scenarios, they are not best for quick thinking and data-acted tasks. Many productivity functions, such as completing transcription or discovering task data, do not require the power of an argument model behind them.
Sara Sachs, Dharna AI Engineering Lead, told venturebeat in an email that the company has targeted for characteristics that do not sacrifice accuracy, security, and privacy, as well as need to respond to Query in speed enterprises.
“To gain low-lonely experience, we corrected the model with internal use and response from trusted examiners, so that AI to make the perception recoveries special,” Sachs said. “This setup assumption helps AI understand business requirements, relevant answers, serve customers with sub-second delay and keep customer data safe and obedient.”
Sachs stated that hosting and construction with various models allow users to “choose the option that meets their needs- whether it is a more conjunctive tone, better coding capabilities, or rapid response time.”
AI meeting notes and more
The perception for the work tracks transcripts meetings for AI and users, especially if they add perception to their calendar, and it can hear on the call.
Users can use perception for enterprise search by adding apps like Slack, Microsoft Teams, Github, Google Drive, SharePoint and Gmail. Sachs stated that Dharna AI would discover an organization’s internal documents, databases and connected apps.
With enterprise search results, other uploaded documents or a web search, perception allows users to reach new research mode. This will “analyze all your sources – along with the web – will draft documents from direct perception to think through the web and through the best response.”
Dharna also added both GPT-4.1 and Cloud 3.7 as a chat option. Openai said that perception can chat with GPT-4.1 on the user workpiece and make a perception template directly from the conversation. Sachs said that the company is working on adding more models to its chat feature.
GPT-4.1 is now made in perception!
You can ask GPT-4.1 to help anything, then create a perception page directly from Convo.
(BTW GPT-4.1 did not have an output in this video.) pic.twitter.com/pvrzpupwrf
– Edwin (@edwinarbus) May 13, 2025
Subscribers for perception business and enterprise plans have immediate access to new features with AI add-on.
Compete with model providers
Even though perception can use both anthropic and openai on the user platform, the perception still has to compete with the model providers.
Openai’s deep research has been seen as a game-chanter for the agent retrieval agmented generation (RAG). Google also has a version of deep research. And anthropic can find internet for you.
To not mention, perception needs to compete with other platforms that already take advantage of AI. Meeting is filled with calls with Space AI to track insight, transcibling, summary and stretching companies.
However, the large sales point of perception is that it has all these abilities on the same platform. Enterprises can use all those various services, but live outside their chosen productivity platform. Dharna said that all these characteristics in one place, with one Sarv-in-One PricingThe enterprises will be prevented from taking membership of separate platforms.
