Kruti AI: Features, Working Model, and Comparison with ChatGPT

• Krutrim launched an agentic AI assistant, Kruti, for real-time tasks like bookings, payments, and shopping.

• Supports voice and text in 13+ languages.

• Conversational AI market: USD 13.9B (2023) → USD 49.9B (2030), CAGR 23.6%.

• The agentic AI market could reach USD 126.9B by 2029, with a CAGR of 35.2%.

• Competes with OpenAI, Google, Microsoft, Anthropic; strong regional alternative.

Kruti is Krutrim's agentic AI assistant that executes real-time tasks like bookings, payments, and shopping through voice and text in 13 Indian languages, competing with ChatGPT, Gemini, and Claude as a regional alternative in the global agentic AI market. Unlike conversational chatbots that only respond to queries, Kruti reasons, plans, and completes multi-step actions autonomously. It supports 13 Indian languages plus global languages, making it accessible to regional users.

Kruti's core distinction from ChatGPT, Gemini, and Claude lies in its agentic framework. When you request "Book me a cab by 8 AM," Kruti checks travel time, books the cab, and sets a reminder without switching applications. This task execution capability defines agentic AI. Traditional chatbots require users to extract information and manually execute actions across different platforms.

The global conversational AI market reached USD 13.9 billion in 2023 and grows at 23.7% annually. The specialized agentic AI market expands faster at 35.2% CAGR, reaching USD 126.9 billion by 2029. Kruti positions itself in this faster-growing segment by combining multilingual support with real-time automation.

The Krutrim AI assistant, Kruti, is currently rolling out primarily in India, reflecting Krutrim’s India-first approach to AI innovation. As an agentic AI assistant, Kruti is being introduced in phases through Krutrim’s ecosystem, with early access expected via mobile apps, web interfaces, and developer APIs.

Krutrim, the AI startup founded by Bhavish Aggarwal, also known for Ola, has indicated that broader public access and enterprise adoption will expand gradually. This phased rollout allows the agentic AI framework behind Kruti to scale responsibly while improving real-time task execution, multilingual accuracy, and reliability.

As global interest in agentic AI assistants grows, Kruti’s rollout strategy positions it as a serious regional alternative to global AI assistants like ChatGPT, Gemini, and Claude, especially for users seeking localized, action-driven AI experiences.

Kruti, Krutrim’s agentic AI assistant, is designed to go beyond traditional chatbots by combining advanced AI models, multi-language support, and intelligent task execution. Its core technology enables it to understand complex requests, plan multi-step actions, and interact seamlessly across languages and platforms. Below is a breakdown of how Kruti achieves this:

1. "Agentic AI" - Not Just a Chatbot

Kruti functions as an "agentic" AI, meaning it can reason, plan, and execute multi-step tasks to fulfill your requests. Instead of just responding with information, it breaks down your request into smaller tasks and completes them.

Example: If you say, "I need to get to the airport by 6 AM," Kruti will:

• Check the time needed to reach the airport

• Book a cab automatically

• Set a reminder for you

• All without you switching between apps

2. Powered by Advanced AI Models

The backend combines Ola's proprietary Krutrim V2 model (with 12 billion parameters) with several open-source large language models. This hybrid approach makes it both powerful and cost-effective.

3. Multi-Language Support

Kruti supports both text and voice in 13 Indian languages, including Hindi, Tamil, Malayalam, and English, with speech recognition built to identify regional dialects and accents.

Example: A user in Kerala can type in Malayalam "എനിക്ക് ഒരു ടാക്സി വേണം" ("I need a taxi") and Kruti will book it and respond in Malayalam.

How Kruti Processes Your Requests

Kruti doesn’t just respond to questions; it acts on them. Its processing framework combines intent recognition, task decomposition, coordinated agents, and direct integration with services to fulfill complex user requests efficiently. Here’s how it works step by step:

Step 1: Understanding Your Intent

Kruti uses intent recognition to parse queries. For example, "Order dosa" triggers the Food Ordering Agent while understanding context like "Make it quick" from previous interactions.

Step 2: Breaking Down Tasks

When a query is submitted, Kruti breaks it into smaller sub-tasks that are carried out via agent-to-agent protocols, reportedly with over 90% accuracy.

Step 3: Multiple Agents Working Together

Kruti uses a network of agents that work together to execute tasks, sometimes in sequence, sometimes in parallel, using an agent-to-agent protocol to coordinate intelligently.

Step 4: Connecting to Real Services

MCP (Model Context Protocol) allows Kruti to communicate directly with databases or APIs, letting it integrate with apps and services to actually complete Botpress.

Two Modes of Operation

Kruti adapts its response style based on the complexity of your requests, offering two distinct modes to ensure efficiency and depth. Whether you need a quick answer or a detailed analysis, Kruti switches seamlessly between Auto Mode and In-Depth Mode:

Auto Mode - For Quick Answers

Auto Mode handles straightforward queries with minimal back-and-forth, like "What's the temperature in Delhi today?" - Kruti responds quickly: "It's 32°C with clear skies."

In-Depth Mode - For Complex Tasks

When users need more than a quick answer, In-Depth Mode steps in, designed for research, learning, or tackling multi-layered questions with detailed, contextual responses.

With agentic AI, trust becomes crucial. Kruti doesn’t just provide answers like ChatGPT, Gemini, or Claude; it acts on them. The system uses confirmations, task checks, and fallback protocols before executing sensitive actions, such as payments or bookings.

While highly capable, Kruti, like all agentic AI frameworks, can face limitations with ambiguous requests. Businesses and users are encouraged to monitor actions, ensuring AI innovation serves them safely.

Kruti handles regional dialects effectively. While ChatGPT, Gemini, and Claude support multiple languages, Kruti’s India-first design ensures accurate real-time execution, even for local phrases like “பர்சல் எப்போது வரும்?” (“When will the parcel arrive?”).

Kruti’s developer SDK and APIs let businesses integrate agentic capabilities into apps:

• E-commerce: automate order tracking & cart recovery

• Travel: manage bookings & itineraries in one conversation

Following approved APIs ensures compliance while leveraging Kruti’s real-time task execution and multilingual support.

Worldwide, the agentic AI market is estimated at USD 28.3 billion in 2024 and expected to grow to USD 126.9 billion by 2029, at a CAGR of 35.2%.

This reflects a key industry evolution:

• North America continues to lead in AI adoption, holding ~38% of global share.

• Asia-Pacific is the fastest-growing region, driven by smartphone penetration and digital ecosystems.

• Europe follows closely with regulatory leadership through the EU AI Act, shaping ethical frameworks for AI deployment.

Krutrim’s localized approach positions it as a regional innovator capable of contributing to this global AI revolution.

Founded by Bhavish Aggarwal, Krutrim builds on India’s AI infrastructure ambitions. The company uses its proprietary large language model (LLM) trained on diverse multilingual datasets, ensuring both accuracy and contextual understanding.

Krutrim has also announced plans for a developer SDK, allowing businesses worldwide to build custom solutions using Kruti’s APIs. This could place Krutrim alongside industry players like Microsoft’s Copilot, Amazon Lex, and Rasa as an emerging competitor in enterprise conversational AI.

Its data sovereignty approach, keeping data localized while enabling global scalability, echoes trends seen in other tech ecosystems prioritizing privacy and national AI frameworks. This focus on integration with CRM and helpdesk systems is becoming crucial for enterprise adoption.

Krutrim has not yet disclosed detailed pricing, but industry insights suggest a hybrid model. Consumers may get basic features of the Krutrim AI assistant free, while businesses and developers could pay for API access, task automation, and enterprise-level integrations.

This model aligns with global generative AI trends, where free access encourages adoption and advanced features drive monetization. It also makes sense for an agentic AI assistant, Kruti, designed to handle real-world tasks like bookings, payments, or generating surreal landscapes on Kruti.ai.

Most comparisons online focus on conversational AI, but Kruti competes in the agentic AI assistant space, not just chatbots. Unlike traditional AI tools, Kruti can reason, plan, and act across multiple steps, a true agentic AI framework.

Users often search for “chatgpt vs kruti comparison”, “kruti vs chat gpt comparison”, or “comparison of Kruti and ChatGPT AI models”. Kruti’s multilingual support, real-time task execution, and regional focus differentiate it from ChatGPT, Gemini, and Claude, highlighting its unique positioning in AI innovation.

Kruti differs from ChatGPT, Gemini, and Claude in three core dimensions: task execution capability, multilingual depth, and regional data sovereignty. The comparison matters because these platforms compete in different segments. ChatGPT excels at conversational AI and content generation. Kruti excels at autonomous task completion and Indian language support.

Real-Time Task Execution: The Core Difference

ChatGPT, Gemini, and Claude respond to prompts with information or generated content. Kruti acts on requests by integrating with external services. When you ask ChatGPT "Book a cab," it explains how to book. When you ask Kruti, it books the cab directly. This distinction separates chatbots from agentic AI.

The table below compares these platforms across their key operational dimensions:

Kruti's real-time execution capability requires direct API integration with payment systems, ride services, and e-commerce platforms. ChatGPT, Gemini, and Claude provide information that users must act on manually or through separate integrations. This architectural difference drives different use cases.

Multilingual Depth and Regional Context

Kruti supports 13 Indian languages including Hindi, Tamil, Telugu, Marathi, Kannada, and Malayalam. ChatGPT supports multiple languages but prioritizes English. Gemini and Claude provide broader language coverage but optimize for global audiences. Kruti optimizes for Indian regional dialects and accents.

Example: A user in Kerala types in Malayalam "എനിക്ക് ഒരു ടാക്സി വേണം" (I need a taxi). Kruti recognizes the intent, processes the regional dialect, and books the taxi in real time, then responds in Malayalam. ChatGPT would translate and explain the booking process. This difference makes Kruti more accessible to non-English speakers seeking direct action, not information.

The voice assistant market reaches USD 14.2 billion by 2032 at 22.89% CAGR. Regional voice assistants grow faster than English-only alternatives because they unlock user bases previously excluded from AI interfaces.

Data Sovereignty and User Control

Kruti emphasizes localized data storage, keeping Indian user data within India. ChatGPT, Gemini, and Claude operate globally with data distributed across multiple jurisdictions. This difference matters for regulatory compliance and user trust.

Organizations in regulated industries such as finance and healthcare often require data to stay within national borders. Kruti's data localization approach enables adoption in these sectors. ChatGPT's global model requires users to manage separate compliance workflows.

Competitive Positioning

Kruti competes as a specialized agentic AI platform for India and Asia-Pacific regions. ChatGPT, Gemini, and Claude compete as general-purpose conversational AI for global audiences. This positioning is not a weakness but a strategy. Krutrim focuses on a specific market segment where it builds deeper integration and trust.

Regional AI platforms like Kruti succeed by understanding local context, regulatory requirements, and user behavior patterns. Global platforms succeed through scale and general-purpose capability. Both approaches create value in different contexts.

Another concern for users asking “what is Kruti?” is how it handles data. Krutrim emphasizes localized storage and user control, especially for Indian users, while meeting global privacy standards.

Kruti’s training for agentic AI capabilities is designed to be transparent, separating user interactions from enterprise or sensitive workflows. This aligns with best practices seen in OpenAI, Anthropic, and Google platforms and ensures a trustworthy AI assistant experience.

Kruti is evolving. Limitations may include phased availability outside India, restricted third-party integrations, and cautious autonomy for high-risk actions.

Despite these, the Krutrim AI assistant Kruti sets the stage for responsible agentic AI, ensuring that users and businesses experience automation safely while leveraging cutting-edge AI innovation.

Kruti’s current focus is India. Phased international rollout planned for Asia-Pacific and Europe, maintaining data sovereignty, multilingual accuracy, and reliable agentic execution.

Kruti and other agentic AI assistants are transforming multiple sectors globally:

• E-commerce: Personalized shopping, cart recovery, order tracking, and generating surreal landscapes on Kruti.ai.

• Healthcare: Patient triage and follow-up assistance.

• Finance: Customer onboarding, KYC automation, and fraud detection.

• Education: Interactive tutoring and adaptive learning.

• Travel: Booking management and real-time itinerary support.

As of 2024, over 73% of enterprises report using AI assistants for customer engagement automation.

Despite rapid innovation, global adoption of agentic AI assistants faces challenges:

• AI Ethics and Bias: Risk of misinformation and discriminatory outputs.

• Regulatory Check: Compliance with the EU AI Act, GDPR, and local privacy laws.

• Data Privacy: Ensuring user control and transparency in personalized AI.

Industry experts highlight that while AI personalization improves experience, it also increases data sensitivity and requires robust governance frameworks. Understanding AI chatbot privacy concerns is essential for businesses deploying these technologies.

For global AI platforms such as Chatboq, Kruti’s emergence highlights the growing need for safe, localized, and compliant conversational AI tools.

Businesses should:

1. Integrate AI responsibly using verified APIs.

2. Prioritize personalization with transparency.

3. Monitor evolving regulations across regions.

4. Experiment with agentic features that automate customer engagement while retaining control.

The next generation of chat platforms will balance automation with ethics, much like Kruti’s approach to responsible AI innovation.

The debut of Krutrim’s Kruti signals a broader transformation in the AI ecosystem, from reactive chatbots to autonomous, personalized assistants that can act on behalf of users.

As generative AI matures, the focus is shifting to trust, localization, and adaptability. Whether in India, the U.S., or Europe, agentic AI assistants like Kruti are redefining how humans and machines collaborate in real-time communication.