Smallest.ai

What Smallest.ai does

Smallest.ai is a full-stack enterprise Voice AI platform founded in 2023 by IIT Guwahati alumni Sudarshan Kamath and Akshat Mandloi. The company develops and operates four proprietary model families — Lightning TTS (sub-100ms latency, 15+ languages), Pulse STT (38 languages with emotion and speaker detection), Electron SLM (sub-3B parameter language model), and Hydra native speech-to-speech — orchestrated through the Atoms agentic platform for end-to-end voice agent deployment. The company targets regulated enterprises in BFSI and healthcare alongside a long tail of SMEs and developers, with named enterprise customers including Paytm and MakeMyTrip and more than 5,000 SME customers across India and the United States.

The company monetizes through a hybrid model combining subscription tiers (Free, $29 Starter, $199 Pro, $1,999+ Enterprise monthly) with usage-based billing at approximately $0.03 per minute of processed voice. Distribution is dual-mode: direct enterprise field sales for regulated verticals and self-serve API access for developers. Compliance certifications (HIPAA, SOC 2 Type 2, ISO 27001, GDPR) qualify the platform for healthcare and financial services deployments. The company reports processing over 10 million calls monthly across 16+ languages.

Smallest.ai is headquartered in San Francisco with engineering operations in Bengaluru and an Indian entity (Awaaz Labs Private Limited) in Pune. The company raised $8 million in seed funding in October 2025 led by Sierra Ventures, with participation from 3one4 Capital, Better Capital, and others, and was selected for the AWS 2025 Generative AI Accelerator. Strategic partnerships include Tenstorrent for hardware-accelerated inference (3.6x cost reduction) and AI Grants India for ecosystem expansion.

Name

Smallest.ai

Legal name

Smallest Inc.

Website

https://smallest.ai

Company type

Private

Founded year

2023

Operating status

Operating

Headcount range

101–250 employees

Short description

Smallest.ai is a full-stack enterprise Voice AI platform offering proprietary TTS, STT, speech-to-speech, and language models for real-time voice agents, serving regulated industries and developers across the US, India, and MENA.

Ownership category

akta.pro rank

Product category

Conversational AI Platform

NAICS

Software Publishers (5132), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (518)

SIC

Services-Prepackaged Software (7372), Services-Computer Programming Services (7371)

akta.pro industry

PRIMARYHDAAAFAC

Conversational AI Platforms (chat/voice bots, orchestration)

Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/Speech, Audio & Conversational AI

SECONDARYHDAEANAJ

AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs)

Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise

SECONDARYHDAAAFAL

Contact Center Speech AI (agent assist, IVR automation)

Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/Speech, Audio & Conversational AI

SECONDARYHDAAACAO

Enterprise Foundation Model Integration & APIs (Connectors, Governance, Deployment)

Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/Foundation Models & Generative AI (LLMs, multimodal)

SECONDARYHDAEANAC

Model Serving, Inference & Deployment Platforms (APIs, Edge/On-Prem)

Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise

Keywords

• Voice AI platform
• Text-to-speech API
• Speech-to-text API
• Conversational AI agents
• Speech-to-speech model

Headquarters

HQ city

San Francisco

HQ country

United States

HQ region

North America

Offices3 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations

Revenue model

1. Voice AI API Platform: SaaS platform revenue generated through API usage fees and subscription plans. Combines recurring subscription components with consumption-based usage billing for API calls and processing minutes.
2. API Usage-Based Billing: Pay-per-usage model where customers are billed based on API call volume and processing minutes consumed. Approximately $0.03 per minute for voice processing.
3. Enterprise Licensing: Custom enterprise agreements with dedicated infrastructure, SLAs, and volume pricing for large-scale deployments.

Pricing tiers

Go-to-market motion3 records

Distribution channels4 records

Marketing channels6 records

Core offering

Smallest.ai develops and operates a full-stack enterprise Voice AI platform that delivers text-to-speech, speech-to-text, small language model, and native speech-to-speech capabilities through a unified API and agent orchestration layer. Enterprises deploy the platform to power autonomous AI voice agents for customer service, lead qualification, debt collection, and transactional call workflows across BFSI, healthcare, retail, real estate, and IT sectors. The offering combines proprietary models with on-premise and cloud deployment options, supported by HIPAA, SOC 2 Type 2, ISO 27001, and GDPR compliance for regulated industries.

Product overview

Smallest.ai is a full-stack enterprise Voice AI platform providing real-time voice automation for contact centers and regulated industries. The platform operates as a unified stack combining Lightning TTS (text-to-speech), Pulse STT (speech-to-text), Electron SLM (small language model), and Hydra speech-to-speech models, orchestrated through the Atoms agentic platform. The product portfolio enables enterprises to deploy AI voice agents for automated customer interactions including support, debt collection, lead qualification, and pre-sales. Key differentiators include sub-100ms TTS latency, multilingual support (15+ languages for TTS, 38 languages for STT), voice cloning, and compliance with HIPAA, GDPR, and SOC 2 for regulated industries. The company offers both cloud-hosted (Hosted Product) and on-premise (Local Product) deployment options.

Differentiator

Problem solved

Functional benefit

Products and services

• Lightning Blazing-fast text-to-speech model designed for real-time voice agents, achieving approximately 100ms latency across 15+ languages. Marketed as production-grade for enterprise voice agent deployments and outperforming OpenAI, Cartesia, and ElevenLabs on voice quality benchmarks. Available via REST API.
• Pulse Speech-to-text transcription API with emotion and speaker detection capabilities available in 38 languages. Optimized for real-time voice agent pipelines with sub-300ms latency, suitable for enterprise transcription and analytics workloads.
• Electron Sub-3 billion parameter small language model that reportedly outperforms GPT-4.1 on speed benchmarks with comparable accuracy, delivering 10x faster time-to-first-byte than GPT-4.1 Mini. Powers conversational intelligence for voice agents via LLM chat completions API.
• Hydra One of the first native speech-to-speech models built for production. Processes continuous audio directly without text intermediate representation, enabling full-duplex conversation with sub-300ms latency, including native handling of interruptions, back-channels, and overlapping speech.
• Atoms Agentic platform for voice agent orchestration. Enables configuration of agents, voice selection, language settings, and deployment from a single interface built for production. Integrates TTS, STT, LLM, and agent logic with HIPAA/GDPR compliance controls.

Quantifiable outcome

• 3.6x cost reduction compared to cloud GPU alternatives
• +3 more outcomes

Named customers2 records

Segments5 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

AI capability8 records

Feature4 records

Partnerships

Three partnerships are on record, tiered core and minor.

• AI Grants IndiacoreStrategic or Co-development Partner · 2 March 2026Collaboration with AI Grants India to provide free Voice AI infrastructure to Indian startups and SMEs. This partnership supports market expansion in India by reducing barriers to adoption for emerging companies and building the voice AI ecosystem in the region.
• TenstorrentcoreTechnology or Integration · 15 March 2025Strategic partnership to deploy Lightning V2 TTS model on Tenstorrent hardware infrastructure. This partnership delivers 3.6x cost reduction compared to cloud GPU alternatives, enabling competitive pricing for real-time voice AI applications. The hardware acceleration partnership is fundamental to Smallest.ai's cost efficiency advantage.
• Amazon Web Services (AWS)minorOthers · 1 January 2025Selected for the AWS 2025 Generative AI Accelerator program, providing cloud credits, technical mentorship, and go-to-market support. This accelerator placement offers visibility, infrastructure benefits, and potential for marketplace listing.

Scale indicators5 records

Recent moves8 records

Expansion highlights7 records

Broad incumbents

• OpenAI: Frontier AI lab whose Realtime API and GPT-4o voice capabilities compete broadly with Smallest.ai's full voice AI stack. While not specialized exclusively in voice, OpenAI's bundled voice features set the benchmark Smallest.ai claims to exceed.
• Google Cloud Speech: Hyperscaler speech-to-text, text-to-speech, and speech-to-speech services (including Gemini Live) that compete with Smallest.ai's Pulse, Lightning, and Hydra models. Bundled into enterprise cloud commitments, creating pricing pressure on standalone voice AI vendors.

Direct peers

• Retell AI: Voice agent platform for building production-grade conversational AI phone agents. Comparable to Smallest.ai's Atoms orchestration layer, targeting similar enterprise use cases in customer support and sales.
• Deepgram: Speech-to-text and voice AI platform overlapping with Smallest.ai's Pulse STT and broader voice agent offering. Both serve enterprise customers with API-first deployment for real-time transcription.
• Play.ht: AI voice generation platform with TTS and voice cloning capabilities. Direct competitor to Smallest.ai's Lightning TTS and Voice Cloning products, targeting content creators and enterprise customers.
• Hume AI: Conversational AI platform focused on emotionally intelligent voice agents with expression measurement. Comparable to Smallest.ai's emotion-detection capabilities in Pulse STT and emotion-aware TTS generation.
• Cartesia: Voice AI foundation model company competing directly with Smallest.ai's Lightning TTS and Pulse STT. Both target real-time voice agent use cases with low-latency model architectures.
• ElevenLabs: Direct competitor in enterprise voice AI, offering TTS, voice cloning, and conversational AI agents. Comparable to Smallest.ai's Lightning TTS and Atoms platform, with overlapping customer targets in enterprise and developer markets.
• Vapi: Voice agent platform enabling developers to build conversational AI phone agents. Direct overlap with Smallest.ai's Atoms platform and Voice Agents offering, serving similar developer and enterprise customer base.
• AssemblyAI: Speech AI platform providing STT, TTS, and audio understanding via API. Overlaps with Smallest.ai's Pulse STT and Lightning TTS, serving similar developer and enterprise customers building voice applications.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat6 records

Key risks6 records

Key highlights6 records

Customer concentration

• Xx.com/smallest_AI
• YouTubeyoutube.com/@smallest_ai
• LinkedInlinkedin.com/company/smallest
• Instagraminstagram.com/smallest.ai

Compliance4 records

Revenue estimate

Valuation estimate

Number of profiles

Profiles3 records

Funding overview

Funding rounds4 records

Investors9 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Smallest.ai