Skip to main content

AVELIN AI

Category: AI Infrastructure

UAE-based sovereign AI platform enabling enterprises and governments in regulated industries to deploy frontier-grade AI with full data ownership, on-premise or air-gapped, via Cross-Model MoE technology. AVELIN AI was founded in 2026. The company is led by Yury Akinin. Based in Dubai, United Arab Emirates. Team size: 11-50. Total funding raised: $3.7M. Latest round: Pre-Seed. Key investors include Angel investors (unnamed).

Founded
2026
Headquarters
Dubai, United Arab Emirates
Team size
11-50
Total funding
$3.7M

Value proposition

Sovereign AI platform combining multiple frontier models (OpenAI, Anthropic, Google, DeepSeek) into a single neural architecture that runs on the customer's own infrastructure — on-premise, air-gapped, or cloud — at 40-60% lower cost than OpenAI/Anthropic, with full data privacy and regulatory compliance.

Products and solutions

AVELIN App (ChatGPT-style sovereign AI workspace), Y-Ray Data Engine (private knowledge base with live web enrichment), API Platform (drop-in OpenAI/Anthropic compatible API for agentic workflows), Cross-Model MoE (9 models across 3 families — chat, coding, agents — with 90% routing accuracy)

Unique value

Only sovereign AI platform offering Cross-Model Mixture-of-Experts (MoE) across entire frontier models (not just internal parameters), delivering higher combined intelligence than any single model, deployable fully inside customer walls — something OpenAI and Anthropic cannot offer.

Target customer

Regulated enterprises in financial services, healthcare, government, and critical infrastructure sectors in the Middle East, Europe, and North America.

Industries served

Financial Services, Healthcare, Government / Public Sector, Critical Infrastructure

Technology advantage

Cross-Model Mixture-of-Experts (MoE) architecture that routes requests across 9 frontier models from 3 families; self-learning routing classifier trained on real usage data; 40-60% cost reduction vs OpenAI/Anthropic; supports cloud, on-premise, and air-gapped deployment; HIPAA, EU AI Act, SOC 2 ready; ISO 27001/27017/27018/27701 certified; new models integrated within 48 hours

How they differentiate

Cross-Model MoE across 9 full frontier models (not just internal parameters) producing higher combined intelligence; available fully on-premise/air-gapped (not just API); 40-60% cheaper than OpenAI/Anthropic; built specifically for regulated Gulf/MEA enterprises with local compliance; integrates both Western (OpenAI, Anthropic, Google) and Eastern (DeepSeek, Qwen) models in a single sovereign stack

Main competitors

Mistral AI (open-weight on-premise models, EU sovereign AI), Cohere (enterprise AI platform with data privacy and compliance), Gaia (sovereign AI platform for regulated enterprises)

Key partnerships

NVIDIA Inception (GPU architecture), AWS Activate (cloud infrastructure), Dubai Future Foundation, Red Hat Partner Program, DigitalOcean Hatch & Partner Pod ISV Track, OVHcloud Startup Program, ElevenLabs (voice synthesis), Lyceum Technology (European GPU infrastructure), Alibaba Cloud, Microsoft, Google Cloud, Hugging Face

Major milestones

Founded 2026 by Yury Akinin, Launched AVELIN GPT (Feb 2026), Joined Red Hat Partner Program (Apr 2026), Secured AWS Infrastructure Backing via Incremental Workloads Program (Apr 2026), Joined OVHcloud Startup Program (Mar 2026), Joined DigitalOcean Hatch (Mar 2026), Advanced to DigitalOcean Partner Pod ISV Track (May 2026), Raised $3.7M Pre-Seed (Jul 2026)

Market positioning

Early-stage UAE sovereign AI platform targeting the rapidly growing Gulf sovereign AI market, competing against larger US/EU foundation model providers by offering on-premise deployment and regulatory compliance as core differentiators for financial services, healthcare, and government customers.

Geographic focus

Middle East (Gulf region), Europe, North America

Patents and IP

Proprietary Cross-Model Fusion technology and Cross-Model MoE architecture (details not publicly disclosed)

About Yury Akinin

Co-Founder and successful exit at Quantori (life sciences AI company, $100M+ valuation); previously led engineering organizations at EPAM Systems and other global enterprises; PhD in Neural Networks from Voronezh State Technical University

Official website: